Compare commits

...

160 Commits

Author SHA1 Message Date
dayuan.jiang
09c556e4c3 chore: bump version to 0.4.1 2025-12-14 23:11:35 +09:00
Dayuan Jiang
ac1c2ce044 fix: remove overly aggressive message filtering on restore (#263)
The hasValidDiagramXml filter was deleting valid messages that had minor
XML issues. Error handling in handleDisplayChart now catches all errors,
so filtering is no longer needed - invalid XML just won't load the diagram
but the conversation is preserved.
2025-12-14 21:49:08 +09:00
Dayuan Jiang
78a77e102d fix: prevent browser crash during long streaming sessions (#262)
- Debounce streaming diagram updates (150ms) to reduce handleDisplayChart calls by 93%
- Debounce localStorage writes (1s) to prevent blocking main thread
- Limit diagramHistory to 20 entries to prevent unbounded memory growth
- Clean up debounce timeout on component unmount to prevent memory leaks
- Add console timing markers for performance profiling

Fixes #78
2025-12-14 21:23:14 +09:00
Dayuan Jiang
55821301dd fix: recover from invalid XML in localStorage on startup (#261)
When LLM generates invalid XML, the app previously saved corrupted messages
to localStorage, causing an unrecoverable crash loop on restart.

This fix validates messages when restoring from localStorage and filters out
any with invalid diagram XML. Users see a toast notification when corrupted
messages are removed.

Fixes #240
2025-12-14 20:01:24 +09:00
Dayuan Jiang
f743219c03 feat: add minimal style mode toggle for faster diagram generation (#260)
* feat: add minimal style mode toggle for faster diagram generation

- Add Minimal/Styled toggle switch in chat input UI
- When enabled, removes color/style instructions from system prompt
- Faster generation with plain black/white diagrams
- Improves XML auto-fix: handle foreign tags, extra closing tags, trailing garbage
- Fix isMxCellXmlComplete to strip Anthropic function-calling wrappers
- Add debug logging for truncation detection diagnosis

* fix: prevent false XML parse errors during streaming

- Escape unescaped & characters in convertToLegalXml() before DOMParser validation
- Only log console.error for final output, not during streaming updates
- Prevents Next.js dev mode error overlay from showing for expected streaming states
2025-12-14 19:38:40 +09:00
Ikko Eltociear Ashimine
ff34f0baf1 docs: update README.md (#257)
Azue -> Azure
2025-12-14 15:08:07 +09:00
Dayuan Jiang
0851b32b67 refactor: simplify LLM XML format to output bare mxCells only (#254)
* refactor: simplify LLM XML format to output bare mxCells only

- Update wrapWithMxFile() to always add root cells (id=0, id=1) automatically
- LLM now generates only mxCell elements starting from id=2 (no wrapper tags)
- Update system prompts and tool descriptions with new format instructions
- Update cached responses to remove root cells and wrapper tags
- Update truncation detection to check for complete mxCell endings
- Update documentation in xml_guide.md

* fix: address PR review issues for XML format refactor

- Fix critical bug: inconsistent truncation check using old </root> pattern
- Fix stale error message referencing </root> tag
- Add isMxCellXmlComplete() helper for consistent truncation detection
- Improve regex patterns to handle any attribute order in root cells
- Update wrapWithMxFile JSDoc to document root cell removal behavior

* fix: handle non-self-closing root cells in wrapWithMxFile regex
2025-12-14 14:04:44 +09:00
Dayuan Jiang
2e24071539 fix: shorten toast notification duration to 2 seconds (#253) 2025-12-14 13:04:18 +09:00
Dayuan Jiang
66bd0e5493 feat: add append_diagram tool and improve truncation handling (#252)
* feat: add append_diagram tool for truncation continuation

When LLM output hits maxOutputTokens mid-generation, instead of
failing with an error loop, the system now:

1. Detects truncation (missing </root> in XML)
2. Stores partial XML and tells LLM to use new append_diagram tool
3. LLM continues generating from where it stopped
4. Fragments are accumulated until XML is complete
5. Server limits to 5 steps via stepCountIs(5)

Key changes:
- Add append_diagram tool definition in route.ts
- Add append_diagram handler in chat-panel.tsx
- Track continuation mode separately from error mode
- Continuation mode has unlimited retries (not counted against limit)
- Error mode still limited to MAX_AUTO_RETRY_COUNT (1)
- Update system prompts to document append_diagram tool

* fix: show friendly message and yellow badge for truncated output

- Add yellow 'Truncated' badge in UI instead of red 'Error' when XML is incomplete
- Show friendly error message for toolUse.input is invalid errors
- Built on top of append_diagram continuation feature

* refactor: remove debug logs and simplify truncation state

- Remove all debug console.log statements
- Remove isContinuationModeRef, derive from partialXmlRef.current.length > 0

* docs: fix append_diagram instructions for consistency

- Change 'Do NOT include' to 'Do NOT start with' (clearer intent)
- Add <mxCell id="0"> to prohibited start patterns
- Change 'closing tags </root></mxGraphModel>' to just '</root>' (wrapWithMxFile handles the rest)
2025-12-14 12:34:34 +09:00
Dayuan Jiang
b33e09be05 feat: add XML auto-fix with refined validation logic (#247)
* feat: add XML auto-fix and improve validator accuracy

- Add autoFixXml() to automatically repair common XML issues:
  - CDATA wrapper removal
  - Duplicate attribute removal
  - Unescaped & and < character escaping
  - Invalid entity reference fixing
  - Unclosed tag completion
  - Nested mxCell flattening
  - Duplicate ID renaming

- Improve validateMxCellStructure() with DOM + regex approach:
  - Use DOMParser for syntax error detection (94% recall)
  - Add regex checks for edge cases
  - Stateful parser for handling > in attribute values

- Integrate validateAndFixXml() in chat-message-display and diagram-context
  - Auto-repair invalid XML before loading
  - Log fixes applied for debugging

Metrics: 99.77% accuracy, 94.06% recall, 94.4% auto-fix success rate

* fix: improve XML auto-fix from 58.7% to 99% fix rate

Key improvements:
- Reorder CDATA removal to run before text-before-root check (+35 cases)
- Implement Gemini's backslash-quote fix with regex backreference
  Handles attr="value", value="text\"inner\"more", and mixed patterns
- Add aggressive drop-broken-cells fix for unfixable mxCell elements
  Iteratively removes cells causing DOM parse errors (up to 50)

Results on 9,411 XML dataset:
- 206 invalid XMLs detected
- 204 successfully fixed (99.0% fix rate)
- 2 unfixable (completely broken, need regeneration)

* refactor: extract XML validation/fix helpers and add constants

- Add constants: MAX_XML_SIZE (1MB), MAX_DROP_ITERATIONS (10), STRUCTURAL_ATTRS, VALID_ENTITIES
- Extract parseXmlTags helper for shared tag parsing logic
- Extract validation helpers: checkDuplicateAttributes, checkDuplicateIds, checkTagMismatches, checkCharacterReferences, checkEntityReferences, checkNestedMxCells
- Simplify validateMxCellStructure from ~200 lines to ~55 lines
- Add logging to empty catch block in DOMParser section
- Add size warning for large XML documents
- Remove unused variables (isSelfClose, duplicate idPattern)

* fix: improve XML auto-fix with malformed quote pattern

- Fix =&quot;...&quot; pattern where &quot; was used as delimiter instead of actual quotes
- Common in dashPattern attributes like dashPattern=&quot;1 1;&quot;
2025-12-13 23:31:01 +09:00
Dayuan Jiang
987dc9f026 fix: add configurable MAX_OUTPUT_TOKENS to prevent truncation (#251)
- Add MAX_OUTPUT_TOKENS env var (fixes output truncation with Bedrock)
- Remove redundant fixToolCallInputs function
- Remove jsonrepair dependency
- Consolidate duplicate lastMessage/userInputText variables
2025-12-13 23:28:41 +09:00
dayuan.jiang
6024443816 fix: improve XML auto-fix from 58.7% to 99% fix rate
Key improvements:
- Reorder CDATA removal to run before text-before-root check (+35 cases)
- Implement Gemini's backslash-quote fix with regex backreference
  Handles attr="value", value="text\"inner\"more", and mixed patterns
- Add aggressive drop-broken-cells fix for unfixable mxCell elements
  Iteratively removes cells causing DOM parse errors (up to 50)

Results on 9,411 XML dataset:
- 206 invalid XMLs detected
- 204 successfully fixed (99.0% fix rate)
- 2 unfixable (completely broken, need regeneration)
2025-12-13 16:11:48 +09:00
dayuan.jiang
4b838fd6d5 feat: add XML auto-fix and improve validator accuracy
- Add autoFixXml() to automatically repair common XML issues:
  - CDATA wrapper removal
  - Duplicate attribute removal
  - Unescaped & and < character escaping
  - Invalid entity reference fixing
  - Unclosed tag completion
  - Nested mxCell flattening
  - Duplicate ID renaming

- Improve validateMxCellStructure() with DOM + regex approach:
  - Use DOMParser for syntax error detection (94% recall)
  - Add regex checks for edge cases
  - Stateful parser for handling > in attribute values

- Integrate validateAndFixXml() in chat-message-display and diagram-context
  - Auto-repair invalid XML before loading
  - Log fixes applied for debugging

Metrics: 99.77% accuracy, 94.06% recall, 94.4% auto-fix success rate
2025-12-13 16:11:47 +09:00
Dayuan Jiang
e321ba7959 chore: optimize Vercel costs by removing analytics and configuring functions (#238)
- Create vercel.json with optimized function settings:
  - Chat API: 512MB memory, 120s timeout
  - Other APIs: 256MB memory, 10s timeout
- Remove @vercel/analytics package and imports
- Reduce chat route maxDuration from 300s to 120s

Expected savings: $2-4/month, keeping costs under $20 included credit
2025-12-12 16:13:06 +09:00
Dayuan Jiang
aa15519fba fix: handle malformed XML from DeepSeek gracefully (#235)
* fix: handle malformed XML from DeepSeek gracefully

Add early XML validation with parsererror check before calling
replaceNodes to prevent application crashes when AI models
generate malformed XML with unescaped special characters.

Changes:
- Add toast import from sonner
- Parse and validate XML before processing
- Add parsererror detection to catch malformed XML early
- Wrap replaceNodes in try-catch for additional safety
- Add user-friendly toast notifications for all error cases
- Change console.log to console.error for validation failures

Fixes #220 #230 #231

* fix: prevent toast spam during streaming and merge silent failure fixes

- Only show error toasts after streaming completes (not during partial updates)
- Track which tool calls have shown errors to prevent duplicate toasts
- Merge clipboard copy error handling from PR #236
- Merge feedback submission error handling from PR #237
- Add comments explaining streaming vs completion behavior

* refactor: simplify toast deduplication with boolean flag

Based on code review feedback, simplified the approach from tracking
per-tool-call IDs in a Set to using a single boolean flag.

Changes:
- Replaced Set<string> with boolean ref for toast tracking
- Removed toolCallId and showToast parameters from handleDisplayChart
- Reset flag when streaming starts (simpler mental model)
- Same behavior: one toast per streaming session, no spam

Benefits:
- Fewer concepts (1 boolean vs Set + 2 parameters)
- No manual coordination between call sites
- Easier to understand and maintain
- ~15 fewer lines of tracking logic

* fix: only show toast for final malformed XML, not during streaming

- Remove errorToastShownRef tracking (no longer needed)
- Add showToast parameter to handleDisplayChart (default false)
- Pass false during streaming (XML may be incomplete)
- Pass true at completion (show toast if final XML is malformed)
- Simpler and more explicit error handling
2025-12-12 14:52:25 +09:00
Dayuan Jiang
c2c65973f9 fix: revert UI and notify user when feedback submission fails (#237)
When feedback submission to the API fails, revert the optimistic
UI update and show a toast notification to inform the user.

Changes:
- Add toast import from sonner
- Change console.warn to console.error for proper logging
- Add toast.error() notification when API call fails
- Revert optimistic UI update by removing feedback from state

Previously, feedback submission failures were completely silent.
Users would see the thumbs-up/down visual feedback but their
feedback was never recorded. This creates a false sense that
the feedback was successfully submitted.

Now users are immediately notified when submission fails and
can retry their feedback.
2025-12-12 14:08:20 +09:00
Dayuan Jiang
b5db980f69 fix: add user feedback for clipboard copy failures (#236)
Add toast notification when clipboard copy operation fails,
so users know when their copy attempt was unsuccessful.

Changes:
- Add toast import from sonner
- Add toast.error() notification when clipboard copy fails
- Show clear message: "Failed to copy message. Please copy
  manually or check clipboard permissions."

Previously, clipboard copy failures were only indicated by a
brief visual state change (setCopyFailedMessageId), which users
might miss. Now users receive persistent feedback when copy
operations fail.
2025-12-12 14:06:53 +09:00
Shashi kiran M S
c9b60bfdb2 feat: Add a new chat button with a confirmation modal (#229)
* feat: Add a new chat button with a confirmation modal

* Fix for PR comments

* fix: add error handling and proper cleanup in handleNewChat

- Add try-catch for localStorage operations to handle quota exceeded,
  private browsing, and other storage errors
- Use handleFileChange([]) instead of setFiles([]) to properly clear
  pdfData Map alongside files
- Only show success toast when localStorage operations succeed
- Show warning toast if localStorage fails but chat state is cleared

---------

Co-authored-by: Dayuan Jiang <jdy.toh@gmail.com>
2025-12-12 10:08:18 +09:00
Twelveeee
f170bb41ae fix:custom model setting bug (#227)
* fix:custom model setting bug

* refactor: consolidate aiProvider checks for cleaner code

---------

Co-authored-by: dayuan.jiang <jdy.toh@gmail.com>
2025-12-12 09:33:07 +09:00
Dayuan Jiang
a0f163fe9e fix: improve Azure provider auto-detection and validation (#223) (#225)
* fix: improve Azure provider auto-detection and validation (#223)

- Fix detectProvider() to only detect Azure when it has complete config
  (both AZURE_API_KEY and AZURE_RESOURCE_NAME or AZURE_BASE_URL)
- Add validation in validateProviderCredentials() for Azure to provide
  clear error messages when configuration is incomplete
- Update docs/ai-providers.md to clarify Azure requires resource name

* docs: add Azure reasoning options to documentation
2025-12-11 21:49:50 +09:00
try2love
8fd3830b9d Fix/clipboard (#189)
* bugfix: clipboard error bug

* fix: use try-catch fallback for clipboard API instead of feature detection

---------

Co-authored-by: dayuan.jiang <jdy.toh@gmail.com>
2025-12-11 21:09:42 +09:00
Biki Kalita
77a25d2543 Persist processed tool calls to prevent replay after chat restore (#224) 2025-12-11 20:48:48 +09:00
Dayuan Jiang
b9da24dd6d fix: limit auto-retry to 3 attempts and enforce quota checks (#219)
- Add MAX_AUTO_RETRY_COUNT (3) to prevent infinite retry loops
- Check token and TPM limits before each auto-retry
- Reset retry counter on user-initiated messages
- Show toast notification when limits are reached

Fixes issue where models returning invalid tool inputs caused 45+ API
requests due to sendAutomaticallyWhen having no retry limit or quota check.
2025-12-11 17:56:40 +09:00
Dayuan Jiang
97cc0a07dc fix: disable history XML replacement by default (#217)
Some models (e.g. minimax) copy placeholder text instead of generating
fresh XML, causing tool call validation failures and infinite loops.

Added ENABLE_HISTORY_XML_REPLACE env var (default: false) to control
this behavior.
2025-12-11 17:36:18 +09:00
dayuan.jiang
c42efdc702 chore: bump version to 0.4.0 and update README features 2025-12-11 14:59:09 +09:00
Dayuan Jiang
dd027f1856 chore: update chain-of-thought.txt example file (#215)
Added full content of the Chain-of-Thought Prompting paper by Wei et al. from Google Research. This example file demonstrates how chain-of-thought reasoning improves LLM performance on complex reasoning tasks.
2025-12-11 14:45:29 +09:00
Dayuan Jiang
869391a029 refactor: eliminate code duplication (DRY principle) (#211)
## Problem Solved
Previous refactoring added 105 lines (1476→1581) by extracting code into separate files without eliminating duplication. This refactor focuses on reducing code size through deduplication while maintaining file separation for maintainability.

## Summary
- Reduced total lines from 1581 to 1519 (-62 lines, 3.9% reduction)
- Eliminated duplicate patterns using generic helpers and factory functions
- Maintained file structure for maintainability
- Zero functional changes - same behavior

### Phase 1: DRY use-quota-manager.tsx
- Created parseStorageCount() helper (eliminates 6x localStorage read duplication)
- Created createQuotaChecker() factory (consolidates 3 check function bodies)
- Created createQuotaIncrementer() factory (consolidates 3 increment function bodies)
- Result: 242→247 lines (+5 lines, but fully DRY with eliminated duplication)

### Phase 2: DRY chat-panel.tsx (1176→1109 lines, -67 lines)

#### 2.1: Extract checkAllQuotaLimits helper
- Replaced 3 occurrences of 18-line quota check blocks
- Saved 36 lines

#### 2.2: Extract sendChatMessage helper
- Replaced 3 occurrences of 21-line sendMessage+headers blocks
- Saved 42 lines

#### 2.3: Extract processFilesAndAppendContent helper
- Replaced 2 occurrences of file processing loops
- Handles PDF, text, and image files uniformly
- Async helper with optional image parts parameter
2025-12-11 14:28:02 +09:00
Dayuan Jiang
8b9336466f feat: make PDF/text extraction char limit configurable via env (#214)
Add NEXT_PUBLIC_MAX_EXTRACTED_CHARS environment variable to allow
configuring the maximum characters extracted from PDF and text files.
Defaults to 150000 (150k chars) if not set.
2025-12-11 14:14:31 +09:00
Dayuan Jiang
ee514efa9e fix: implement AZURE_RESOURCE_NAME config for Azure OpenAI (#213)
Previously AZURE_RESOURCE_NAME was documented in env.example but not
actually used in the code. This caused Azure OpenAI configuration to fail
when users set AZURE_RESOURCE_NAME instead of AZURE_BASE_URL.

Changes:
- Read AZURE_RESOURCE_NAME from environment and pass to createAzure()
- resourceName constructs endpoint: https://{name}.openai.azure.com/openai/v1
- baseURL takes precedence over resourceName when both are set
- Updated env.example with clearer documentation

Fixes #208
2025-12-11 13:32:33 +09:00
Dayuan Jiang
e2757a34b7 Replace old asset link with new one
Updated asset link in README.md.
2025-12-11 12:25:00 +09:00
Dayuan Jiang
c0347dd55d fix: prevent diagram regeneration loop after successful display (#206)
- Changed sendAutomaticallyWhen to only auto-resubmit on tool errors
- Extracted logic to shouldAutoResubmit() function with JSDoc
- Added TypeScript interfaces for type safety (MessagePart, ChatMessage)
- Wrapped debug logs with DEBUG flag for production readiness
- Added TOOL_ERROR_STATE constant to avoid hardcoded strings

Problem: AI was regenerating diagrams 3+ times even after successful display
Root cause: lastAssistantMessageIsCompleteWithToolCalls auto-resubmits on both success AND error
Solution: Custom logic that only auto-resubmits on errors, stops on success
2025-12-11 09:47:30 +09:00
Biki Kalita
a047a6ff97 feat: Display AI reasoning/thinking blocks in chat interface (#152)
* feat: Add reasoning/thinking blocks display in chat interface

* feat: add multi-provider options support and replace custom reasoning UI with AI Elements

* resolve conflicting reasoning configs and correct provider-specific reasoning parameters

* try to solve conflict

* fix: simplify reasoning display and remove unnecessary dependencies

- Remove Streamdown dependency (~5MB) - reasoning is plain text only
- Fix Bedrock providerOptions merging for Claude reasoning configs
- Remove unsupported DeepSeek reasoning configuration
- Clean up unused environment variables (REASONING_BUDGET_TOKENS, REASONING_EFFORT, DEEPSEEK_REASONING_*)
- Remove dead commented code from route.ts

Reasoning blocks contain plain thinking text and don't need markdown/diagram/code rendering.

* feat: comprehensive reasoning support improvements

Major improvements:
- Auto-enable reasoning display for all supported models
- Fix provider-specific reasoning configurations
- Remove unnecessary Streamdown dependency (~5MB)
- Clean up debug logging

Provider changes:
- OpenAI: Auto-enable reasoningSummary for o1/o3/gpt-5 models
- Google: Auto-enable includeThoughts for Gemini 2.5/3 models
- Bedrock: Restrict reasoningConfig to only Claude/Nova (fixes MiniMax error)
- Ollama: Add thinking support for qwen3-like models

Other improvements:
- Remove ENABLE_REASONING toggle (always enabled)
- Fix Bedrock providerOptions merging for Claude
- Simplify reasoning component (plain text rendering)
- Clean up unused environment variables

* fix: critical bugs and documentation gaps in reasoning support

Critical fixes:
- Fix Bedrock shallow merge bug (deep merge preserves anthropicBeta + reasoningConfig)
- Add parseInt validation with parseIntSafe helper (prevents NaN errors)
- Validate all numeric env vars with min/max ranges

Documentation improvements:
- Add BEDROCK_REASONING_BUDGET_TOKENS and BEDROCK_REASONING_EFFORT to env.example
- Add OLLAMA_ENABLE_THINKING to env.example
- Update JSDoc with accurate env var list and ranges

Code cleanup:
- Remove debug console.log statements from route.ts
- Refactor duplicate providerOptions assignments

---------

Co-authored-by: Dayuan Jiang <34411969+DayuanJiang@users.noreply.github.com>
Co-authored-by: Dayuan Jiang <jdy.toh@gmail.com>
2025-12-11 00:24:43 +09:00
Dayuan Jiang
d2ba133eaf feat: add PDF and text file upload support (#205)
- Add client-side PDF text extraction using unpdf library
- Support text files (.txt, .md, .json, .csv, .py, .js, .ts, etc.)
- Add file preview with character count for PDF/text files
- Add 150k character limit for extracted content
- Highlight Paper to Diagram example with NEW badge
- Fix React hydration error by adding explicit IDs to ResizablePanelGroup
- Remove code duplication by centralizing file utilities in pdf-utils.ts
2025-12-10 21:32:35 +09:00
Dayuan Jiang
43e5993f47 fix: improve LLM diagram context awareness and image preview (#202)
- Add replaceHistoricalToolInputs to replace XML in tool calls with placeholders
- Send both previousXml and current xml so LLM can understand user's manual edits
- Update system message to mark current XML as authoritative source of truth
- Fix React StrictMode issue with blob URL cleanup in FilePreviewList
- Add unoptimized prop to Image components for blob URLs
2025-12-10 18:04:37 +09:00
Dayuan Jiang
9a954ccb44 fix: correct NEXT_PUBLIC_DRAWIO_BASE_URL in offline deployment docs (#198)
- Fix wrong URL (http://drawio:8080 -> http://localhost:8080)
- Browser cannot resolve Docker internal hostnames
- Add docker-compose.yml example
- Simplify documentation
2025-12-10 14:20:34 +09:00
Dayuan Jiang
9d4c89ec43 docs: improve offline deployment guide with docker compose (#191) 2025-12-10 09:40:48 +09:00
try2love
5da4ef67ec feat:light/dark mode switch (#138)
Summary

- Adds browser theme detection on first visit using
prefers-color-scheme media query
- Renames localStorage key from dark-mode to
next-ai-draw-io-dark-mode for consistency with other keys
- Uses STORAGE_DIAGRAM_XML_KEY constant instead of hardcoded
string in diagram-context.tsx

Changes

app/page.tsx:
- On first visit (no saved preference), detect browser's color
scheme preference
- Update localStorage key to follow project naming convention
(next-ai-draw-io-*)

contexts/diagram-context.tsx:
- Import STORAGE_DIAGRAM_XML_KEY from chat-panel.tsx
- Replace hardcoded "next-ai-draw-io-diagram-xml" with the
constant
2025-12-10 09:21:15 +09:00
Dayuan Jiang
67b0adf211 docs: align Chinese and Japanese README with main README (#190) 2025-12-10 08:42:53 +09:00
Dayuan Jiang
65af353852 Update model requirements and notes in README 2025-12-10 08:19:21 +09:00
Dayuan Jiang
b3deb65584 chore: remove duplicate self-hosted-drawio.md (#188)
Content has been consolidated into docs/offline-deployment.md
2025-12-09 22:27:34 +09:00
Dayuan Jiang
626f3a76b5 docs: add offline deployment guide (#187) 2025-12-09 22:19:28 +09:00
terrydash
97bb350dd6 feat: add support for self-hosted draw.io via DRAWIO_BASE_URL environment variable (#163)
## Summary
Add support for self-hosted draw.io instances via build-time configuration.

## Problem
In some corporate environments, `embed.diagrams.net` is blocked by network policies. 
Users cannot use the application without access to the default draw.io embed URL.

## Solution
- Add `NEXT_PUBLIC_DRAWIO_BASE_URL` environment variable support
- Pass the `baseUrl` prop to the `DrawIoEmbed` component
- Configure Dockerfile to accept build-time argument for the draw.io URL

## Usage
```yaml
# docker-compose.yaml
services:
  drawio:
    image: jgraph/drawio:latest
    ports:
      - "8080:8080"
  
  next-ai-draw-io:
    build:
      context: .
      args:
        - NEXT_PUBLIC_DRAWIO_BASE_URL=http://drawio:8080
```

Or build directly:
```bash
docker build --build-arg NEXT_PUBLIC_DRAWIO_BASE_URL=http://localhost:8080 -t next-ai-draw-io .
```

**Note:** This is a build-time configuration. To change the draw.io URL, you need to rebuild the Docker image.
2025-12-09 22:00:54 +09:00
Dayuan Jiang
97ab82e027 feat: add bring-your-own-API-key support (#186)
- Add AI provider settings to config panel (provider, model, API key, base URL)
- Support 7 providers: OpenAI, Anthropic, Google, Azure, OpenRouter, DeepSeek, SiliconFlow
- Client API keys stored in localStorage, never stored on server
- Client settings override server env vars when provided
- Skip server credential validation when client provides API key
- Bypass usage limits (request/token/TPM) when using own API key
- Add /api/config endpoint for fetching usage limits
- Add privacy notices to settings dialog, about pages, and quota toast
- Add clear settings button to reset saved API keys
- Update README files (EN/CN/JA) with BYOK documentation

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-09 17:50:07 +09:00
dayuan.jiang
77cb10393b fix: translate image content block error to user-friendly message 2025-12-09 16:00:19 +09:00
Dayuan Jiang
967d63c57e feat: support minimax model (#185)
* feat: support minimax model with XML wrapping fix

- Add wrapWithMxFile utility to properly wrap XML for draw.io
- Fix 'Not a diagram file' error when model generates raw <root> XML
- Add supportsPromptCaching check for conditional caching
- Only enable Bedrock prompt caching for Claude models

* docs: update model mention to minimax-m2 across About pages and READMEs

- Update tooltip in chat-panel.tsx to mention minimax-m2 model change
- Update English, Chinese, and Japanese About pages with model change info
- Update English, Chinese, and Japanese READMEs with demo site model note

---------

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-09 15:53:59 +09:00
Dayuan Jiang
914e914423 feat: replace hardcoded usage limits with dynamic env variables (#180)
- About pages now read DAILY_REQUEST_LIMIT, DAILY_TOKEN_LIMIT, TPM_LIMIT from env
- Removed unused /app/api/config/ route
- Numbers formatted as Xk (e.g., 30k, 10k)
2025-12-09 09:57:09 +09:00
Dayuan Jiang
f6cfcab45a fix: upgrade React to 19.1.2 to patch CVE-2025-55182 (#176)
Fixes critical RCE vulnerability (CVSS 10.0) in React Server Components
caused by deserialization of untrusted data.

Closes #175
2025-12-08 23:31:17 +09:00
singledog957
95c5a75ca3 feat: Show detailed error messages instead of generic 'Internal server error' (#144) (#154)
* feat: Show detailed error messages instead of generic 'Internal server error' (#144)

* refactor: simplify error handling logic per feedback

* refactor: imported AI SDK error handler

* fix: remove unused import and expand sensitive data filter

- Remove unused NoSuchModelError import
- Add 'secret', 'password', 'credential' to sensitive data filter

---------

Co-authored-by: dayuan.jiang <jdy.toh@gmail.com>
2025-12-08 20:52:18 +09:00
Dayuan Jiang
ac09f9f8f9 feat: redesign usage limits section on about pages (#173)
- Redesign usage limits card with gradient border and modern styling
- Remove emojis and combine title/subtitle on same line
- Make all 3 language pages (EN/CN/JP) consistent in design
- Update text content with exact localized wording
- Add warning triangle icon in chat panel linking to about page
- Add 'Learn more' link in quota limit toast
- Open about page links in new tab to preserve diagram state
2025-12-08 20:26:51 +09:00
Dayuan Jiang
622829b903 feat: add daily token limit with actual usage tracking (#171)
* feat: add daily token limit with actual usage tracking

- Add DAILY_TOKEN_LIMIT env var for configurable daily token limit
- Track actual tokens from Bedrock API response metadata (not estimates)
- Server sends inputTokens + cachedInputTokens + outputTokens via messageMetadata
- Client increments token count in onFinish callback with actual usage
- Add NaN guards to prevent corrupted localStorage values
- Add token limit toast notification with quota display
- Remove client-side token estimation (was blocking legitimate requests)
- Switch to js-tiktoken for client compatibility (pure JS, no WASM)

* feat: add TPM (tokens per minute) rate limiting

- Add 50k tokens/min client-side rate limit
- Track tokens per minute with automatic minute rollover
- Check TPM limit after daily limits pass
- Show toast when rate limit reached
- NaN guards for localStorage values

* feat: make TPM limit configurable via TPM_LIMIT env var

* chore: restore cache debug logs

* fix: prevent race condition in TPM tracking

checkTPMLimit was resetting TPM count to 0 when checking, which
overwrote the count saved by incrementTPMCount. Now checkTPMLimit
only reads and incrementTPMCount handles all writes.

* chore: improve TPM limit error message clarity
2025-12-08 18:56:34 +09:00
Dayuan Jiang
728dda5267 feat: add daily request limit with custom toast notification (#167)
- Add DAILY_REQUEST_LIMIT env var support in config API
- Track request count in localStorage (resets daily)
- Show friendly quota limit toast with self-host/sponsor links
- Apply limit to send, regenerate, and edit message actions
2025-12-08 14:26:01 +09:00
Dayuan Jiang
b68b1b0f33 chore: clean up root directory by moving docs (#166)
Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-08 12:08:35 +09:00
Dayuan Jiang
bd23aed93b docs: update README with badges, TOC, and live demo button (#165)
* docs: update README with badges, TOC, and reorganized sections

* docs: add SVG button for live demo link
2025-12-08 11:48:14 +09:00
Dayuan Jiang
95aa4b8a56 chore: remove Amplify integration (#164)
Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-08 11:39:32 +09:00
Dayuan Jiang
4070772733 perf: disable prefetch on About link to reduce requests (#162) 2025-12-08 11:03:29 +09:00
Dayuan Jiang
c4aaa7c915 feat: only show access code field when password is configured (#161)
Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-08 11:00:14 +09:00
Dayuan Jiang
ecea8a6005 fix: use static maxDuration value for Next.js 16 compatibility (#160) 2025-12-08 10:56:37 +09:00
Dayuan Jiang
ebc622144b chore: add Cloudflare-related files to gitignore (#159) 2025-12-08 10:42:08 +09:00
Dayuan Jiang
ee9267d54c chore: make maxDuration configurable via env variable (#157)
Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-08 10:20:52 +09:00
Dayuan Jiang
f6682fe3ac fix(diagram): wrap XML in mxfile structure when canvas is empty (#156)
When clicking an example diagram on a clean page (no existing diagram),
the code was passing raw <root>...</root> format XML directly to DrawIO.
However, DrawIO expects the full mxfile/diagram/mxGraphModel/root structure.

This fix provides a default mxfile template when chartXML is empty,
ensuring replaceNodes properly wraps the content before loading into DrawIO.

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-08 10:10:52 +09:00
Biki Kalita
03db4c8096 fix(diagram): save history snapshot after edit_diagram tool execution (#155) 2025-12-08 09:44:10 +09:00
dayuan.jiang
167f5ed36a feat: enable recordInputs in Langfuse telemetry
Enable full message history recording including XML tool calls for better observability.
2025-12-07 20:58:44 +09:00
Dayuan Jiang
cd8e0e2263 feat: add token counting utility for system prompts (#153)
Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 20:33:43 +09:00
Dayuan Jiang
8c431ee6ed fix: preserve message parts order in chat display (#151)
- Fix bug where text after tool calls was merged with initial text
- Group consecutive text/file parts into bubbles while keeping tools in order
- Parts now display as: plan -> tool_result -> additional text
- Remove debug logs from fixToolCallInputs function

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 19:56:31 +09:00
Dayuan Jiang
86420a42c6 fix: implement client-side caching for example diagrams (#150)
- Add client-side cache check in onFormSubmit to bypass API calls for example prompts
- Use findCachedResponse to match input against cached examples
- Directly set messages with cached tool response when example matches
- Hide regenerate button for cached example responses (toolCallId starts with 'cached-')
- Prevents unnecessary API calls when using example buttons

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 19:36:09 +09:00
Dayuan Jiang
0baf21fadb fix: validate XML before displaying diagram to catch duplicate IDs (#147)
- Add validation to loadDiagram in diagram-context, returns error or null
- display_diagram and edit_diagram tools now check validation result
- Return error to AI agent with state: output-error so it can retry
- Skip validation for trusted sources (localStorage, history, internal templates)
- Add debug logging for tool call inputs to diagnose Bedrock API issues
2025-12-07 14:38:15 +09:00
Dayuan Jiang
a54068fec2 docs: add --env-file option for Docker and simplify install instructions (#142) 2025-12-07 11:49:01 +09:00
Dayuan Jiang
e25fd367d5 chore: add GitHub issue templates for bug reports and feature requests (#140) 2025-12-07 11:00:25 +09:00
Aurelius Huang
3264244fe9 fix: add deps: @opentelemetry/exporter-trace-otlp-http version 0.208.0 (#124) 2025-12-07 10:49:01 +09:00
QiyuanChen
d8cdd049d1 feat: add SiliconFlow as a supported AI provider (#137)
* feat: add SiliconFlow as a supported AI provider in documentation and configuration

* fix: update SiliconFlow configuration comment to English
2025-12-07 10:22:57 +09:00
Dayuan Jiang
b1bc1a6dc6 feat: auto-save and restore session state (#135)
- Save and restore chat messages, XML snapshots, session ID, and diagram XML to localStorage
- Restore diagram when DrawIO becomes ready (using new onLoad callback)
- Change close protection default to false since auto-save handles persistence
- Clear localStorage when clearing chat
- Handle edge cases: undefined edit fields, empty chartXML, missing access code header
2025-12-07 01:39:09 +09:00
Biki Kalita
8b578a456e fix: Remove hardcoded temperature parameter to support models that don't support it (#133)
* Fix: remove hardcoded temperature parameter to support reasoning models

* feat: make temperature configurable via AI_TEMPERATURE env var

- Instead of removing temperature entirely, make it optional via env var
- Set AI_TEMPERATURE=0 for deterministic output (recommended for diagrams)
- Leave unset for models that don't support temperature (e.g., GPT-5.1 reasoning)

* docs: add AI_TEMPERATURE env var documentation

- Update env.example with AI_TEMPERATURE option
- Update README.md configuration section
- Add Temperature Setting section in ai-providers.md

* docs: add TEMPERATURE env var documentation

- Update env.example with TEMPERATURE option
- Update README.md, README_CN.md, README_JA.md configuration sections
- Add Temperature Setting section in ai-providers.md
- Update route.ts to use TEMPERATURE env var

---------

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 01:34:59 +09:00
Dayuan Jiang
05d58025c4 fix: fix hydration mismatch for DrawIO theme loading (#131)
- Load DrawIO theme from localStorage after mount with useEffect
- Add loading spinner while theme loads
- Prevents SSR/client hydration mismatch (server has no localStorage)

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 00:45:19 +09:00
Dayuan Jiang
4be64317b3 feat: enhance system prompts with JSON escaping and edge routing rules (#132)
- Add JSON escaping warnings to help model generate valid tool calls
- Add comprehensive edge routing rules to prevent overlapping lines
- Add planning guidance for diagram creation
- Update token count estimates in comments

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 00:40:23 +09:00
Dayuan Jiang
2fac6323f0 fix: add orphaned mxPoint validation and cleanup (#130)
- Add validation for orphaned mxPoint elements in validateMxCellStructure()
- Add cleanup of orphaned mxPoint elements in convertToLegalXml()
- Orphaned mxPoints cause 'Could not add object mxPoint' errors in draw.io
- mxPoint elements must have 'as' attribute or be inside <Array as="points">

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 00:40:19 +09:00
Dayuan Jiang
a415c46b66 feat: improve XML search/replace matching strategies (#129)
- Add 6th strategy: match by value attribute (label text)
- Add 7th strategy: normalized whitespace match
- Remove lastProcessedIndex tracking - always search from beginning
- Pairs may not be in document order, so sequential tracking was unreliable

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 00:40:16 +09:00
Dayuan Jiang
3894abd9ed feat: add tool call JSON repair and Bedrock compatibility (#127)
- Add fixToolCallInputs() to fix Bedrock API requirement (JSON object, not string)
- Add experimental_repairToolCall for malformed JSON from model
- Add stepCountIs(5) limit to prevent infinite loops
- Update edit_diagram tool description with JSON escaping warning

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 00:40:13 +09:00
Dayuan Jiang
6965a54f48 feat: upgrade @ai-sdk/react to 2.0.107 and migrate to new API (#126)
- Upgrade @ai-sdk/react from 2.0.22 to 2.0.107
- Migrate from addToolResult to addToolOutput (new API)
- Add output-error state for proper error signaling to model
- Add sendAutomaticallyWhen for auto-retry on tool errors
- Add stop function ref for potential future use

Co-authored-by: dayuan.jiang <jiangdy@amazon.co.jp>
2025-12-07 00:40:11 +09:00
Dayuan Jiang
46567cb0b8 feat: verify access code with server before saving (#128) 2025-12-07 00:21:59 +09:00
Dayuan Jiang
9f77199272 feat: add configurable close protection setting (#123)
- Add Close Protection toggle to Settings dialog
- Save setting to localStorage (default: enabled)
- Make beforeunload confirmation conditional
- Settings button now always visible in header
- Add shadcn Switch and Label components
2025-12-06 21:42:28 +09:00
dayuan.jiang
77f2569a3b chore: bump version to 0.3.0 2025-12-06 19:26:26 +09:00
Dayuan Jiang
cbb92bd636 fix: set maxDuration to 60 for Vercel hobby plan (#122) 2025-12-06 18:09:30 +09:00
Dayuan Jiang
8d898d8adc fix: revert maxDuration to static value (Next.js requirement) (#121) 2025-12-06 18:04:23 +09:00
Dayuan Jiang
1e0b1ed970 feat: make maxDuration configurable via MAX_DURATION env (#120) 2025-12-06 17:47:50 +09:00
Dayuan Jiang
1d03d10ba8 docs: add CONTRIBUTING.md (#119) 2025-12-06 17:39:47 +09:00
Dayuan Jiang
e893bd60f9 fix: resolve biome lint errors and memory leak in file preview (#118)
- Disable noisy biome rules (noExplicitAny, useExhaustiveDependencies, etc.)
- Fix memory leak in file-preview-list.tsx with useRef pattern
- Separate unmount cleanup into dedicated useEffect
- Add ToolPartLike interface for type safety in chat-message-display
- Add accessibility attributes (role, tabIndex, onKeyDown)
- Replace autoFocus with useEffect focus pattern
- Minor syntax improvements (optional chaining, key fixes)
2025-12-06 16:18:26 +09:00
Dayuan Jiang
9aaf9bf31f refactor: deduplicate system prompts with two-phase composition (#117) 2025-12-06 12:58:53 +09:00
Dayuan Jiang
150eb1ff63 chore: add Biome for formatting and linting (#116)
- Add Biome as formatter and linter (replaces Prettier)
- Configure Husky + lint-staged for pre-commit hooks
- Add VS Code settings for format on save
- Ignore components/ui/ (shadcn generated code)
- Remove semicolons, use 4-space indent
- Reformat all files to new style
2025-12-06 12:46:40 +09:00
Dayuan Jiang
215a101f54 fix: revert edit_diagram tool description to original (#115) 2025-12-06 12:41:01 +09:00
Dayuan Jiang
e00938d9d3 feat: enhance system prompt with app context and dynamic model name (#114)
- Add App Context section describing the left/right panel layout
- Add App Features section with icon locations (history, theme, upload, export, clear)
- Dynamically inject model name into system prompt via {{MODEL_NAME}} placeholder
- Expand edit_diagram tool description with usage guidelines
2025-12-06 12:37:37 +09:00
Dayuan Jiang
dd27d034e2 fix: force re-render when switching between mobile/desktop layout (#112) 2025-12-05 23:45:57 +09:00
Dayuan Jiang
9e781005af fix: make button hover state darker instead of lighter (#111) 2025-12-05 23:38:24 +09:00
Dayuan Jiang
fe1aa2747e fix: add viewport meta tag for mobile layout (#110) 2025-12-05 23:34:18 +09:00
Dayuan Jiang
7205896c8c feat: add mobile layout with chat panel at bottom (#109) 2025-12-05 23:25:59 +09:00
Dayuan Jiang
4e32a094b1 fix: restore status notice indicator removed in PR #77 (#107) 2025-12-05 23:22:29 +09:00
Dayuan Jiang
96a1111654 fix: ensure markdown text in user messages is visible (#108)
The prose plugin was overriding text colors for markdown elements
(bold, headings, etc.) in user message bubbles, causing text to
blend with the dark primary background.

Added conditional styling that forces all child elements in user
messages to use text-primary-foreground color with !important to
override prose defaults.
2025-12-05 23:16:59 +09:00
Dayuan Jiang
3f35c52527 feat: add draw.io theme toggle between minimal and sketch (#106)
- Add toggle button in chat input area to switch between min and sketch themes
- Show warning dialog before switching (clears messages and diagram)
- Persist theme selection in localStorage
- Default theme is minimal (hides shapes sidebar)
2025-12-05 23:10:48 +09:00
Dayuan Jiang
0af5229477 feat: add markdown rendering and resizable chat panel (#104)
* feat: add markdown rendering for chat messages

- Add react-markdown and @tailwindcss/typography for markdown support
- Use prose styling for assistant message formatting
- Fix Radix ScrollArea viewport horizontal overflow issue
- Add CSS fix for viewport width constraint

* feat: add resizable chat panel

- Replace fixed width layout with react-resizable-panels
- Chat panel can be resized by dragging the handle
- Panel is collapsible with min 15% and max 50% width
- Ctrl+B keyboard shortcut still works for toggle
2025-12-05 22:42:39 +09:00
Twelveeee
3fb349fb3e clear button cant clear error msg & feat: add setting dialog and add accesscode (#77)
* fix: clear button cant clear error msg

* new: add setting dialog and add accesscode

* fix: address review feedback - dark mode, types, formatting

* feat: only show Settings button when access code is required

* refactor: rename ACCESS_CODES to ACCESS_CODE_LIST

---------

Co-authored-by: dayuan.jiang <jdy.toh@gmail.com>
2025-12-05 22:09:34 +09:00
Dayuan Jiang
ed29e32ba3 feat: restore Langfuse observability integration (#103)
- Add lib/langfuse.ts with client, trace input/output, telemetry config
- Add instrumentation.ts for OpenTelemetry setup with Langfuse span processor
- Add /api/log-save endpoint for logging diagram saves
- Add /api/log-feedback endpoint for thumbs up/down feedback
- Update chat route with sessionId tracking and telemetry
- Add feedback buttons (thumbs up/down) to chat messages
- Add sessionId tracking throughout the app
- Update env.example with Langfuse configuration
- Add @langfuse/client, @langfuse/otel, @langfuse/tracing, @opentelemetry/sdk-trace-node
2025-12-05 21:15:02 +09:00
Dayuan Jiang
4cd78dc561 chore: remove complex 503 error handling code (#102)
- Remove 15s streaming timeout detection (too slow, added complexity)
- Remove status indicator (issue resolved by switching model)
- Remove streamingError state and related refs
- Simplify onFinish callback (remove 503 detection logging)
- Remove errorHandler function (use default AI SDK errors)

The real fix was switching from global.* to us.* Bedrock model.
This removes ~134 lines of unnecessary complexity.
2025-12-05 20:18:19 +09:00
Dayuan Jiang
e0c5d966e3 feat: add image upload validation with 2MB limit and max 5 files (#101)
- Add 2MB file size limit with client and server-side validation
- Add max 5 files limit per upload
- Add sonner toast library for better error notifications
- Create ErrorToast component with keyboard accessibility
- Batch multiple validation errors into single toast
- Validate file size in all upload methods (input, paste, drag-drop)
- Add server-side validation in /api/chat endpoint
2025-12-05 19:30:50 +09:00
Dayuan Jiang
33471d5b3a docs: add AI provider configuration guide (#100)
- Add docs/ai-providers.md with detailed setup instructions for all providers
- Update README.md, README_CN.md, README_JA.md with provider guide links
- Add model capability requirements note
- Simplify provider list in READMEs

Closes #79
2025-12-05 18:53:34 +09:00
Dayuan Jiang
3ef9908df7 feat: add confirmation dialog to prevent accidental back navigation (#99)
Addresses conflict between right-click drag and browser back gesture in
Chromium-based browsers. Shows browser confirmation dialog when user
tries to navigate away, preventing accidental page exits.

Closes #80
2025-12-05 18:42:36 +09:00
Dayuan Jiang
57bfc9cef7 fix: update status indicator to show outage resolved (#98) 2025-12-05 18:07:25 +09:00
Dayuan Jiang
0543f71c43 fix: use console.log instead of console.error for XML validation during streaming (#96) 2025-12-05 16:59:14 +09:00
Dayuan Jiang
970b88612d fix: add service status indicator for ongoing issues (#95) 2025-12-05 16:46:17 +09:00
Dayuan Jiang
c805277a76 fix: enable UI retry when Bedrock returns early 503 error (#94)
- Add error prop to ChatInput to detect error state
- Update isDisabled logic to allow retry when there's an error
- Pass combined error (SDK error + streamingError) to ChatInput

When Bedrock returns 503 ServiceUnavailableException before streaming
starts, AI SDK's onError fires but status may not transition to "ready".
This fix ensures the input is re-enabled when an error occurs, allowing
users to retry their request.
2025-12-05 16:22:38 +09:00
Dayuan Jiang
95160f5a21 fix: handle Bedrock 503 streaming errors with timeout detection (#92)
- Add 15s streaming timeout to detect mid-stream stalls (e.g., Bedrock 503)
- Add stop() call to allow user retry after timeout
- Add streamingError state for timeout-detected errors
- Improve server-side error logging for empty usage detection
- Add user-friendly error messages for ServiceUnavailable and Throttling errors
2025-12-05 14:23:47 +09:00
broBinChen
b206e16c02 fix: clear files when clicking text-only examples (#82)
Fixed an issue where files from previous examples would persist when clicking on "Animated Diagram" or "Creative Drawing" examples that don't require image uploads.
2025-12-05 14:07:14 +09:00
broBinChen
563b18e8ff refactor: replace deprecated addToolResult with addToolOutput (#85)
Replaced the deprecated addToolResult API with the new addToolOutput API from ai to ensure compatibility with future versions.
2025-12-05 14:02:45 +09:00
dayuan.jiang
2366255e8f fix: use credential provider chain for bedrock IAM role support 2025-12-05 09:19:26 +09:00
dayuan.jiang
255308f829 fix: make bedrock credentials optional for IAM role support 2025-12-05 09:11:10 +09:00
dayuan.jiang
a9493c8877 fix: write env vars to .env.production for Amplify SSR runtime 2025-12-05 09:04:54 +09:00
dayuan.jiang
a0c3db100a fix: add favicon.ico to public folder for header logo 2025-12-05 08:56:34 +09:00
dayuan.jiang
ff6f130f8a refactor: remove Langfuse observability integration
- Delete lib/langfuse.ts, instrumentation.ts
- Remove API routes: log-save, log-feedback
- Remove feedback buttons (thumbs up/down) from chat
- Remove sessionId tracking throughout codebase
- Remove @langfuse/*, @opentelemetry dependencies
- Clean up env.example
2025-12-05 01:30:02 +09:00
dayuan.jiang
562751c913 fix: disable recordInputs to prevent Langfuse media upload timeout
When images are included in chat messages, the AI SDK telemetry with
recordInputs: true sends base64 image data to Langfuse. Langfuse then
attempts to upload these images to media storage, causing 1m31s timeouts.

Setting recordInputs: false prevents this while still capturing user
text input via setTraceInput().
2025-12-05 01:14:01 +09:00
dayuan.jiang
95e8a9c0c0 fix: update chartXMLRef directly before sendMessage to avoid race condition
The React state update (setChartXML) is async, so chartXMLRef wasn't updated
when edit_diagram tool callback checked it. Now we update the ref directly
in onFormSubmit, handleRegenerate, and handleEditMessage before sending.
2025-12-05 00:54:35 +09:00
dayuan.jiang
d9568562f0 fix: use ref for chartXML to avoid stale closure in onToolCall
The onToolCall callback was capturing stale chartXML value due to
JavaScript closure. Using a ref ensures we always get the latest value.
2025-12-05 00:47:27 +09:00
dayuan.jiang
7b8bd8c621 fix: use cached chartXML for edit_diagram to avoid Vercel timeout
DrawIO iframe export was unreliable on Vercel due to network latency,
causing edit_diagram tool to hang. Now uses chartXML from context directly,
falling back to export only when no cached XML exists.
2025-12-05 00:43:21 +09:00
dayuan.jiang
46cbc3354c fix: add manual token usage reporting to Langfuse for Bedrock streaming
Bedrock streaming responses don't auto-report token usage to OpenTelemetry.
This fix manually sets span attributes (ai.usage.promptTokens, gen_ai.usage.input_tokens)
from the AI SDK onFinish callback to ensure Langfuse captures token counts.
2025-12-05 00:26:02 +09:00
dayuan.jiang
46d2d4e078 refactor: add input validation and singleton pattern for Langfuse API routes
- Add Zod schema validation for log-feedback and log-save endpoints
- Create singleton LangfuseClient to avoid per-request instantiation
- Simplify log-save to only flag trace (no XML content sent)
- Use generic error messages to prevent info leakage
2025-12-04 23:44:00 +09:00
dayuan.jiang
d8f2c85dab feat: link user feedback and diagram saves to chat traces in Langfuse
- Update log-feedback API to find existing chat trace by sessionId and attach score to it
- Update log-save API to create span on existing chat trace instead of standalone trace
- Add thumbs up/down feedback buttons on assistant messages
- Add message regeneration and edit functionality
- Add save dialog with format selection (drawio, png, svg)
- Pass sessionId through components for Langfuse linking
2025-12-04 22:56:59 +09:00
Dayuan Jiang
5f4d31e708 fix: auto-detect AI provider from configured API keys (#74)
- Remove default bedrock provider requirement
- Auto-detect provider when only one API key is configured
- Show helpful error when no keys or multiple keys without AI_PROVIDER
- Fixes #73
2025-12-04 14:13:10 +09:00
Dayuan Jiang
489b377063 chore: upgrade Next.js from 15.2.3 to 16.0.7 (#72)
- Fixes critical CVE-2025-66478 (CVSS 10.0) - RSC protocol vulnerability
- Includes Turbopack stability improvements
- Updated tsconfig.json with Next.js 16 recommended settings
2025-12-04 13:48:30 +09:00
Dayuan Jiang
3534cb13f7 refactor: extract system prompts and add extended prompt for Opus/Haiku 4.5 (#71)
- Extract system prompts to dedicated lib/system-prompts.ts module
- Add extended system prompt (~4000 tokens) for models with higher cache minimums (Opus 4.5, Haiku 4.5)
- Clean up debug logs while preserving informational and cache-related logs
- Improve code formatting and organization in chat route
2025-12-04 13:26:06 +09:00
Dayuan Jiang
9d9613a8d1 feat: add trace-level input/output to Langfuse observability (#69)
* feat: add trace-level input/output to Langfuse observability

- Add @langfuse/client and @langfuse/tracing dependencies
- Wrap POST handler with observe() for proper tracing
- Use updateActiveTrace() to set trace input, output, sessionId, userId
- Filter Next.js HTTP spans in shouldExportSpan so AI SDK spans become root traces
- Enable recordInputs/recordOutputs in experimental_telemetry

* refactor: extract Langfuse logic to separate lib/langfuse.ts module
2025-12-04 11:24:26 +09:00
Dayuan Jiang
bed04c82f8 chore: add Apache 2.0 license and update gitignore (#68)
* feat: integrate Langfuse for LLM observability

- Add instrumentation.ts with Langfuse OpenTelemetry exporter
- Enable experimental telemetry on streamText calls
- Add instrumentationHook to Next.js config
- Install required dependencies (@vercel/otel, langfuse-vercel, etc.)

* feat: add optional Langfuse observability integration

- Add session tracking with unique sessionId per conversation
- Add user tracking via IP address (x-forwarded-for header)
- Make telemetry conditional - only enabled if LANGFUSE_PUBLIC_KEY is set
- Add environment variable validation in instrumentation.ts
- Add sessionId validation (type check + 200 char limit)
- Update env.example with Langfuse configuration docs
- Remove unused langfuse-vercel and @vercel/otel packages

* fix: remove deprecated instrumentationHook (enabled by default in Next.js 15)

* chore: add Apache 2.0 license and update gitignore
2025-12-04 00:33:32 +09:00
Dayuan Jiang
fa1b02ad78 feat: integrate Langfuse for LLM observability (#66)
* feat: integrate Langfuse for LLM observability

- Add instrumentation.ts with Langfuse OpenTelemetry exporter
- Enable experimental telemetry on streamText calls
- Add instrumentationHook to Next.js config
- Install required dependencies (@vercel/otel, langfuse-vercel, etc.)

* feat: add optional Langfuse observability integration

- Add session tracking with unique sessionId per conversation
- Add user tracking via IP address (x-forwarded-for header)
- Make telemetry conditional - only enabled if LANGFUSE_PUBLIC_KEY is set
- Add environment variable validation in instrumentation.ts
- Add sessionId validation (type check + 200 char limit)
- Update env.example with Langfuse configuration docs
- Remove unused langfuse-vercel and @vercel/otel packages

* fix: remove deprecated instrumentationHook (enabled by default in Next.js 15)
2025-12-04 00:23:09 +09:00
Dayuan Jiang
39322c2793 fix: prevent duplicate history entries when edit_diagram tool is called (#64)
- Add handleExportWithoutHistory function for fetching current diagram state without saving to history
- Update onFetchChart to accept saveToHistory parameter (defaults to true)
- edit_diagram tool now fetches with saveToHistory=false since it only needs the current state
- Only the initial form submission saves to history as intended
2025-12-03 21:58:48 +09:00
Dayuan Jiang
110cccb09c feat: refresh UI with new typography and edit diff display (#63)
- Switch from Geist to Plus Jakarta Sans (body) and JetBrains Mono (code)
- Add visual diff display for edit_diagram tool showing search/replace pairs
- Update color palette to clean modern OKLCH-based scheme
- Improve chat message display with better styling and animations
- Add syntax-highlighted code blocks for XML/JSON output
- Improve scrollbar and shadow utilities
2025-12-03 21:49:34 +09:00
Dayuan Jiang
5021076864 fix: use static sha- prefix for Docker tags and add Docker run instructions to README (#62) 2025-12-03 21:15:25 +09:00
Dayuan Jiang
efdf4f2b90 chore: change clear button icon to trash bin (#61) 2025-12-03 21:05:22 +09:00
Dayuan Jiang
45f74df349 feat: add save diagram to local file button (#60)
- Add save button in chat input area with download icon
- Create SaveDialog component for filename input
- Export current diagram as .drawio file format
- Support custom filename with default timestamp-based name

Closes #53
2025-12-03 21:02:26 +09:00
Dayuan Jiang
a61d37c818 chore: add Google Analytics integration (#59) 2025-12-03 20:35:31 +09:00
Dayuan Jiang
c0cd393baa fix: validate XML before sending to draw.io to prevent setId errors (#56)
- Add XML validation in handleDisplayChart before calling onDisplayChart
- Only update previousXML ref when validation passes to prevent state desync
- Add console error logging for failed validations

Fixes #5
2025-12-03 20:11:50 +09:00
Dayuan Jiang
595f24857a fix: show user-friendly error when model doesn't support images (#55)
When models like DeepSeek (deepseek-chat, deepseek-reasoner) receive image
inputs, they return a cryptic error about 'unknown variant image_url'.
This change detects such errors and shows a clear message asking users
to remove the image or switch to a vision-capable model.

Fixes #42
2025-12-03 19:49:58 +09:00
Dayuan Jiang
33fed6fa9f chore: add model switch notice and move sponsor button to about page (#52)
- Remove sponsor iframe from chat panel header
- Add notice about switching from Opus 4.5 to Haiku 4.5 due to high traffic
- Add sponsor button next to Support & Contact section title
- Update all i18n about pages (EN, CN, JA)
2025-12-03 16:47:45 +09:00
Dayuan Jiang
a8e627f1f8 feat: add XML structure guide to system prompt for smaller models (#51)
- Add essential draw.io XML structure rules to system prompt
- Include critical rules about mxCell nesting (all must be direct children of root)
- Add shape/vertex and connector/edge examples with proper structure
- Improve tool description for display_diagram with validation rules
- Update xml_guide.md with better swimlane examples showing flat structure
- Add client-side XML validation to catch nested mxCell errors early

Helps address issues #40 (local Ollama models not working) and #39 (mxCell nesting errors)
2025-12-03 16:14:53 +09:00
Dayuan Jiang
c458947553 feat: add confirmation dialog for diagram history restore (#49) 2025-12-03 14:04:29 +09:00
Dayuan Jiang
443a937370 fix: prevent duplicate diagram history entries on message send (#48)
When sending a message, the history was being added twice because:
1. handleExport() triggers exportDiagram() which adds to history
2. AI responds and loadDiagram() is called, which internally triggers
   another export event in DrawIO, adding a duplicate entry

Added expectHistoryExportRef flag to track user-initiated exports and
only add to history when the export was explicitly requested.
2025-12-03 13:53:16 +09:00
Dayuan Jiang
3f5cdd807d fix: prevent tool args from expanding chat width during streaming (#47) 2025-12-03 13:45:14 +09:00
Dayuan Jiang
894740ba58 feat: add i18n about pages matching README structure (#46)
- Redesign English about page to mirror README.md content
- Add Chinese (/about/cn) and Japanese (/about/ja) versions
- Include language switcher, features, examples with images
- Add multi-provider support section and contact info
2025-12-03 13:36:36 +09:00
Dayuan Jiang
271f3b0f58 Merge pull request #44 from jianyun8023/dev
Add Docker Support with Multi-stage Build
2025-12-03 09:02:23 +09:00
dayuan.jiang
bc0f767ad7 Merge main into dev, resolve conflicts 2025-12-03 08:59:46 +09:00
dayuan.jiang
61ef41addf docs: add Docker badge to READMEs 2025-12-03 08:56:16 +09:00
Jianyun8023
5d38ed59eb feat: add Docker support with multi-stage build 2025-12-02 21:58:06 +08:00
Dayuan Jiang
53754e627a feat: add GitHub Sponsors button and i18n READMEs (#43)
- Add GitHub Sponsors iframe button to chat panel header
- Update README with badges and language switcher
- Add Chinese README (README_CN.md)
- Add Japanese README (README_JA.md)
- Reorganize examples section in README
2025-12-02 22:53:31 +09:00
Dayuan Jiang
bca80c0856 Update GitHub Sponsors username in FUNDING.yml 2025-12-02 22:03:19 +09:00
Dayuan Jiang
e2adfb49aa Merge pull request #38 from DayuanJiang/feat/add-deepseek-provider
feat: add DeepSeek as AI provider
2025-12-02 11:59:35 +09:00
dayuan.jiang
45ab934288 feat: add DeepSeek as AI provider
- Install @ai-sdk/deepseek package
- Add DeepSeek provider support to lib/ai-providers.ts
- Add DeepSeek configuration to env.example
- Update README.md with DeepSeek in provider list
- Support both default and custom base URL for DeepSeek
2025-12-02 11:52:09 +09:00
Dayuan Jiang
af3173623a Merge pull request #36 from DayuanJiang/chore/remove-github-workflows
chore: remove github workflows
2025-12-02 01:19:45 +09:00
dayuan.jiang
cd012f5e2f chore: remove github workflows 2025-12-02 01:12:49 +09:00
Dan Zheng
d4fb635d98 fix: add customize anthropic baseURL (#28)
* fix: add custom anthropic baseURL

* feat: add baseURL support for all AI providers

- Add GOOGLE_BASE_URL for Google Generative AI
- Add AZURE_BASE_URL for Azure OpenAI
- Add OLLAMA_BASE_URL support (was documented but not implemented)
- Add OPENROUTER_BASE_URL for OpenRouter
- Fix missing semicolon in Anthropic case
- Update env.example with new environment variables

Closes #20

---------

Co-authored-by: dayuan.jiang <jdy.toh@gmail.com>
2025-12-02 01:08:06 +09:00
Dayuan Jiang
14740e35a8 Update README with bug fix and star history section 2025-12-01 22:40:43 +09:00
Dayuan Jiang
5b31216917 feat: cache example prompt responses to save tokens (#34)
- Add lib/cached-responses.ts with pre-generated XML for 4 example prompts
- Modify chat API route to check cache before calling AI
- Cache returns instant response (~0.26s) vs AI generation (~20-25s)
- Add "(cached for instant response)" text to example panel
- Cache only activates for first message with empty diagram
2025-12-01 14:07:50 +09:00
Dayuan Jiang
c7d0260328 feat: add Bedrock prompt caching for system and conversation messages (#32)
* feat: add Bedrock prompt caching for system and conversation messages

- Add cache point to system message (2558+ tokens cached)
- Add cache point to last assistant message in conversation history
- This caches the entire conversation prefix for subsequent requests
- Reduces latency and costs for multi-turn conversations

* refactor: remove duplicated system prompt
2025-12-01 10:43:33 +09:00
Dayuan Jiang
d2d4dd01cc fix: filter out messages with empty content arrays for Bedrock API (#31)
* fix: filter out messages with empty content arrays for Bedrock API

The convertToModelMessages function from AI SDK can produce messages with
empty content arrays when assistant messages have only tool call parts or
when tool results aren't properly converted. Bedrock API rejects these with
400 errors. This fix filters out invalid messages before sending to the API.

* fix: add diagnostic logging for empty message content

Added logging to capture the original UI message structure when empty content
is detected after conversion. This helps debug the root cause while the
filter provides a safety net for Bedrock API compatibility.
2025-12-01 01:15:43 +09:00
Dayuan Jiang
b4679f6598 fix: increase maxDuration to 300s for Fluid Compute (#30) 2025-12-01 00:46:40 +09:00
Dayuan Jiang
0d0d553e23 fix: correct anthropic beta header config for fine-grained tool streaming (#27)
* fix: correct anthropic beta header config for fine-grained tool streaming

- Use bedrock.anthropicBeta for Bedrock provider (not additionalModelRequestFields)
- Use top-level headers for direct Anthropic API
- Update @ai-sdk/amazon-bedrock to 3.0.62
- Add headers support to ModelConfig interface

* fix: update @ai-sdk/amazon-bedrock to 3.0.62 for tool streaming support
2025-11-30 16:34:42 +09:00
81 changed files with 17553 additions and 2463 deletions

60
.dockerignore Normal file
View File

@@ -0,0 +1,60 @@
# Dependencies
node_modules
npm-debug.log*
yarn-debug.log*
yarn-error.log*
# Build output
.next
out
dist
build
# Testing
coverage
.nyc_output
# Environment variables
.env
.env*.local
.env.local
.env.development.local
.env.test.local
.env.production.local
# Git
.git
.gitignore
.gitattributes
# IDE
.vscode
.idea
*.swp
*.swo
*~
# Operating System
.DS_Store
Thumbs.db
# Documentation
README.md
*.md
!env.example
# CI/CD
.github
.gitlab-ci.yml
.travis.yml
# Docker
Dockerfile
.dockerignore
docker-compose*.yml
# Other
*.log
.cache
.turbo

View File

@@ -1,6 +1,3 @@
{
"extends": [
"next/core-web-vitals",
"next/typescript"
]
"extends": ["next/core-web-vitals", "next/typescript"]
}

35
.github/CONTRIBUTING.md vendored Normal file
View File

@@ -0,0 +1,35 @@
# Contributing
## Setup
```bash
git clone https://github.com/YOUR_USERNAME/next-ai-draw-io.git
cd next-ai-draw-io
npm install
cp env.example .env.local
npm run dev
```
## Code Style
We use [Biome](https://biomejs.dev/) for linting and formatting:
```bash
npm run format # Format code
npm run lint # Check lint errors
npm run check # Run all checks (CI)
```
Pre-commit hooks via Husky will run Biome automatically on staged files.
For a better experience, install the [Biome VS Code extension](https://marketplace.visualstudio.com/items?itemName=biomejs.biome) for real-time linting and format-on-save.
## Pull Requests
1. Create a feature branch
2. Make changes and ensure `npm run check` passes
3. Submit PR against `main` with a clear description
## Issues
Include steps to reproduce, expected vs actual behavior, and AI provider used.

15
.github/FUNDING.yml vendored Normal file
View File

@@ -0,0 +1,15 @@
# These are supported funding model platforms
github: dayuanjiang
patreon: # Replace with a single Patreon username
open_collective: # Replace with a single Open Collective username
ko_fi: # Replace with a single Ko-fi username
tidelift: # Replace with a single Tidelift platform-name/package-name e.g., npm/babel
community_bridge: # Replace with a single Community Bridge project-name e.g., cloud-foundry
liberapay: # Replace with a single Liberapay username
issuehunt: # Replace with a single IssueHunt username
lfx_crowdfunding: # Replace with a single LFX Crowdfunding project-name e.g., cloud-foundry
polar: # Replace with a single Polar username
buy_me_a_coffee: # Replace with a single Buy Me a Coffee username
thanks_dev: # Replace with a single thanks.dev username
custom: # Replace with up to 4 custom sponsorship URLs e.g., ['link1', 'link2']

35
.github/ISSUE_TEMPLATE/bug_report.md vendored Normal file
View File

@@ -0,0 +1,35 @@
---
name: Bug Report
about: Report a bug to help us improve
title: '[Bug] '
labels: bug
assignees: ''
---
> **Note**: This template is just a guide. Feel free to ignore the format entirely - any feedback is welcome! Don't let the template stop you from sharing your thoughts.
## Bug Description
A brief description of the issue.
## Steps to Reproduce
1. Go to '...'
2. Click on '...'
3. Scroll to '...'
4. See error
## Expected Behavior
What you expected to happen.
## Actual Behavior
What actually happened.
## Screenshots
If applicable, add screenshots to help explain the problem.
## Environment
- OS: [e.g. Windows 11, macOS 14]
- Browser: [e.g. Chrome 120, Safari 17]
- Version: [e.g. 1.0.0]
## Additional Context
Any other information about the problem.

5
.github/ISSUE_TEMPLATE/config.yml vendored Normal file
View File

@@ -0,0 +1,5 @@
blank_issues_enabled: true
contact_links:
- name: Discussions
url: https://github.com/DayuanJiang/next-ai-draw-io/discussions
about: Have questions or ideas? Feel free to start a discussion

View File

@@ -0,0 +1,25 @@
---
name: Feature Request
about: Suggest a new feature for this project
title: '[Feature] '
labels: enhancement
assignees: ''
---
> **Note**: This template is just a guide. Feel free to ignore the format entirely - any feedback is welcome! Don't let the template stop you from sharing your ideas.
## Feature Description
A brief description of the feature you'd like.
## Problem Context
Is this related to a problem? Please describe.
e.g. I'm always frustrated when [...]
## Proposed Solution
How you'd like this feature to work.
## Alternatives Considered
Any alternative solutions or features you've considered.
## Additional Context
Any other information or screenshots about the feature request.

View File

@@ -1,118 +0,0 @@
name: Claude Code
on:
issue_comment:
types: [created]
pull_request_review_comment:
types: [created]
issues:
types: [opened, assigned]
pull_request_review:
types: [submitted]
pull_request_target:
types: [opened, synchronize, reopened]
jobs:
claude:
if: |
(github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||
(github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||
(github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||
(github.event_name == 'issues' && contains(github.event.issue.body, '@claude'))
runs-on: ubuntu-latest
permissions:
contents: write
pull-requests: write
issues: write
id-token: write
steps:
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Configure AWS Credentials
uses: aws-actions/configure-aws-credentials@v4
with:
role-to-assume: ${{ secrets.AWS_ROLE_TO_ASSUME }}
aws-region: us-east-1
- name: Run Claude Code
uses: anthropics/claude-code-action@v1
with:
use_bedrock: "true"
github_token: ${{ secrets.GITHUB_TOKEN }}
pr-review:
if: github.event_name == 'pull_request_target'
runs-on: ubuntu-latest
permissions:
contents: read
pull-requests: write
id-token: write
steps:
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Configure AWS Credentials
uses: aws-actions/configure-aws-credentials@v4
with:
role-to-assume: ${{ secrets.AWS_ROLE_TO_ASSUME }}
aws-region: us-east-1
- name: Run Claude Code PR Review
uses: anthropics/claude-code-action@v1
with:
use_bedrock: "true"
github_token: ${{ secrets.GITHUB_TOKEN }}
allowed_non_write_users: "*"
prompt: |
REPO: ${{ github.repository }}
PR NUMBER: ${{ github.event.pull_request.number }}
This is a personal project - an AI-powered draw.io diagram generator built with:
- Next.js 15 with React 19
- Vercel AI SDK (streamText, useChat, tool calling)
- Multiple AI providers: Bedrock, Anthropic, OpenAI, Google, Azure, OpenRouter, Ollama
STEP 1: Check existing comments to avoid duplicates.
Run: `gh api repos/${{ github.repository }}/pulls/${{ github.event.pull_request.number }}/comments`
Build a list of files and line numbers that already have comments. For each existing comment:
- If the issue is FIXED in current code, resolve the thread using:
`gh api graphql -f query='mutation { resolveReviewThread(input: {threadId: "THREAD_ID"}) { thread { isResolved } } }'`
- If the issue still exists, remember this file:line - DO NOT create a new comment for it
STEP 2: Review the diff for issues, but SKIP any file:line that already has a comment.
Review this PR for these issues (report ALL that apply):
1. Bugs that would cause runtime errors or broken functionality
2. Security issues (exposed secrets, API key leaks)
3. AI SDK misuse - specifically check for:
- Client-side: Should use useChat/useCompletion/useObject hooks, NOT raw fetch()
- Server-side: Should use streamText/generateText/streamObject/generateObject
- Message handling: Access message.parts array, not legacy content property
- Tool definitions: Must use Zod schemas for inputSchema
- Status handling: Check status (submitted/streaming/ready/error) before actions
- Stream cleanup: Call stop() when aborting streams
4. Unrelated changes that should be in separate PRs (scope creep)
5. Suspicious .gitignore additions or accidentally committed files
6. UI/UX inconsistencies (e.g., alignment issues)
When reviewing AI SDK usage, fetch https://ai-sdk.dev/docs/ to verify correct patterns.
Key doc pages: /docs/ai-sdk-ui/chatbot, /docs/ai-sdk-core/generating-text, /docs/ai-sdk-core/tools-and-tool-calling
DO NOT comment on:
- Minor performance optimizations
- Code style preferences (unless clearly wrong)
- Type annotations that don't affect functionality
IMPORTANT:
- NEVER create a comment on a file:line that already has a comment - this causes duplicates
- For each NEW issue, use `mcp__github_inline_comment__create_inline_comment` to comment on the specific line
- ALWAYS include a suggested fix using GitHub's suggestion syntax: ```suggestion\n<fixed code>\n```
- Only say "LGTM" if there are truly ZERO new issues to report
claude_args: |
--allowedTools "mcp__github_inline_comment__create_inline_comment,Bash(gh pr diff:*),Bash(gh pr view:*),Bash(gh api:*),WebFetch(domain:ai-sdk.dev)"

66
.github/workflows/docker-build.yml vendored Normal file
View File

@@ -0,0 +1,66 @@
name: Docker Build and Push
on:
push:
branches:
- main
- master
- dev
tags:
- 'v*'
pull_request:
branches:
- main
- master
env:
REGISTRY: ghcr.io
IMAGE_NAME: ${{ github.repository }}
jobs:
build-and-push:
runs-on: ubuntu-latest
permissions:
contents: read
packages: write
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Log in to GitHub Container Registry
if: github.event_name != 'pull_request'
uses: docker/login-action@v3
with:
registry: ${{ env.REGISTRY }}
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
- name: Extract metadata (tags, labels)
id: meta
uses: docker/metadata-action@v5
with:
images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
tags: |
type=ref,event=branch
type=ref,event=pr
type=semver,pattern={{version}}
type=semver,pattern={{major}}.{{minor}}
type=semver,pattern={{major}}
type=sha,prefix=sha-
type=raw,value=latest,enable={{is_default_branch}}
- name: Build and push Docker image
uses: docker/build-push-action@v5
with:
context: .
push: ${{ github.event_name != 'pull_request' }}
tags: ${{ steps.meta.outputs.tags }}
labels: ${{ steps.meta.outputs.labels }}
cache-from: type=gha
cache-to: type=gha,mode=max
platforms: linux/amd64,linux/arm64

7
.gitignore vendored
View File

@@ -40,4 +40,9 @@ yarn-error.log*
*.tsbuildinfo
next-env.d.ts
push-via-ec2.sh
.claude/settings.local.json
.claude/
.playwright-mcp/
# Cloudflare
.dev.vars
.open-next/
.wrangler/

1
.husky/pre-commit Normal file
View File

@@ -0,0 +1 @@
npx lint-staged

23
.vscode/settings.json vendored Normal file
View File

@@ -0,0 +1,23 @@
{
"editor.formatOnSave": true,
"editor.defaultFormatter": "biomejs.biome",
"editor.codeActionsOnSave": {
"source.fixAll.biome": "explicit",
"source.organizeImports.biome": "explicit"
},
"[javascript]": {
"editor.defaultFormatter": "biomejs.biome"
},
"[typescript]": {
"editor.defaultFormatter": "biomejs.biome"
},
"[javascriptreact]": {
"editor.defaultFormatter": "biomejs.biome"
},
"[typescriptreact]": {
"editor.defaultFormatter": "biomejs.biome"
},
"[json]": {
"editor.defaultFormatter": "biomejs.biome"
}
}

59
Dockerfile Normal file
View File

@@ -0,0 +1,59 @@
# Multi-stage Dockerfile for Next.js
# Stage 1: Install dependencies
FROM node:20-alpine AS deps
RUN apk add --no-cache libc6-compat
WORKDIR /app
# Copy package files
COPY package.json package-lock.json* ./
# Install dependencies
RUN npm ci
# Stage 2: Build application
FROM node:20-alpine AS builder
WORKDIR /app
# Copy node_modules from deps stage
COPY --from=deps /app/node_modules ./node_modules
COPY . .
# Disable Next.js telemetry during build
ENV NEXT_TELEMETRY_DISABLED=1
# Build-time argument for self-hosted draw.io URL
ARG NEXT_PUBLIC_DRAWIO_BASE_URL=https://embed.diagrams.net
ENV NEXT_PUBLIC_DRAWIO_BASE_URL=${NEXT_PUBLIC_DRAWIO_BASE_URL}
# Build Next.js application (standalone mode)
RUN npm run build
# Stage 3: Production runtime
FROM node:20-alpine AS runner
WORKDIR /app
ENV NODE_ENV=production
ENV NEXT_TELEMETRY_DISABLED=1
# Create non-root user for security
RUN addgroup --system --gid 1001 nodejs
RUN adduser --system --uid 1001 nextjs
# Copy necessary files
COPY --from=builder /app/public ./public
# Copy standalone build output
COPY --from=builder --chown=nextjs:nodejs /app/.next/standalone ./
COPY --from=builder --chown=nextjs:nodejs /app/.next/static ./.next/static
USER nextjs
EXPOSE 3000
ENV PORT=3000
ENV HOSTNAME="0.0.0.0"
# Start the application
CMD ["node", "server.js"]

190
LICENSE Normal file
View File

@@ -0,0 +1,190 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to the Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
Copyright 2024 Dayuan Jiang
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

206
README.md
View File

@@ -1,26 +1,59 @@
# Next AI Draw.io
A next.js web application that integrates AI capabilities with draw.io diagrams. This app allows you to create, modify, and enhance diagrams through natural language commands and AI-assisted visualization.
<div align="center">
https://github.com/user-attachments/assets/b2eef5f3-b335-4e71-a755-dc2e80931979
**AI-Powered Diagram Creation Tool - Chat, Draw, Visualize**
Demo site: [https://next-ai-draw-io.vercel.app](https://next-ai-draw-io.vercel.app)
English | [中文](./docs/README_CN.md) | [日本語](./docs/README_JA.md)
## Features
[![TrendShift](https://trendshift.io/api/badge/repositories/15449)](https://next-ai-drawio.jiang.jp/)
- **LLM-Powered Diagram Creation**: Leverage Large Language Models to create and manipulate draw.io diagrams directly through natural language commands
- **Image-Based Diagram Replication**: Upload existing diagrams or images and have the AI replicate and enhance them automatically
- **Diagram History**: Comprehensive version control that tracks all changes, allowing you to view and restore previous versions of your diagrams before the AI editing.
- **Interactive Chat Interface**: Communicate with AI to refine your diagrams in real-time
- **AWS Architecture Diagram Support**: Specialized support for generating AWS architecture diagrams
- **Animated Connectors**: Create dynamic and animated connectors between diagram elements for better visualization
[![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
[![Next.js](https://img.shields.io/badge/Next.js-16.x-black)](https://nextjs.org/)
[![React](https://img.shields.io/badge/React-19.x-61dafb)](https://react.dev/)
[![Sponsor](https://img.shields.io/badge/Sponsor-❤-ea4aaa)](https://github.com/sponsors/DayuanJiang)
## **Examples**
[![Live Demo](./public/live-demo-button.svg)](https://next-ai-drawio.jiang.jp/)
</div>
A Next.js web application that integrates AI capabilities with draw.io diagrams. Create, modify, and enhance diagrams through natural language commands and AI-assisted visualization.
https://github.com/user-attachments/assets/9d60a3e8-4a1c-4b5e-acbb-26af2d3eabd1
## Table of Contents
- [Next AI Draw.io ](#next-ai-drawio-)
- [Table of Contents](#table-of-contents)
- [Examples](#examples)
- [Features](#features)
- [Getting Started](#getting-started)
- [Try it Online](#try-it-online)
- [Run with Docker (Recommended)](#run-with-docker-recommended)
- [Installation](#installation)
- [Deployment](#deployment)
- [Multi-Provider Support](#multi-provider-support)
- [How It Works](#how-it-works)
- [Project Structure](#project-structure)
- [Support \& Contact](#support--contact)
- [Star History](#star-history)
## Examples
Here are some example prompts and their generated diagrams:
<div align="center">
<table width="100%">
<tr>
<td colspan="2" valign="top" align="center">
<strong>Animated transformer connectors</strong><br />
<p><strong>Prompt:</strong> Give me a **animated connector** diagram of transformer's architecture.</p>
<img src="./public/animated_connectors.svg" alt="Transformer Architecture with Animated Connectors" width="480" />
</td>
</tr>
<tr>
<td width="50%" valign="top">
<strong>GCP architecture diagram</strong><br />
@@ -40,44 +73,67 @@ Here are some example prompts and their generated diagrams:
<img src="./public/azure_demo.svg" alt="Azure Architecture Diagram" width="480" />
</td>
<td width="50%" valign="top">
<strong>Animated transformer connectors</strong><br />
<p><strong>Prompt:</strong> Give me a **animated connector** diagram of transformer's architecture.</p>
<img src="./public/animated_connectors.svg" alt="Transformer Architecture with Animated Connectors" width="480" />
</td>
</tr>
<tr>
<td colspan="2" valign="top" align="center">
<strong>Cat sketch prompt</strong><br />
<p><strong>Prompt:</strong> Draw a cute cat for me.</p>
<img src="./public/cat_demo.svg" alt="Cat Drawing" width="260" />
<img src="./public/cat_demo.svg" alt="Cat Drawing" width="240" />
</td>
</tr>
</table>
</div>
## How It Works
## Features
The application uses the following technologies:
- **Next.js**: For the frontend framework and routing
- **@ai-sdk/react**: For the chat interface and AI interactions
- **react-drawio**: For diagram representation and manipulation
Diagrams are represented as XML that can be rendered in draw.io. The AI processes your commands and generates or modifies this XML accordingly.
## Multi-Provider Support
- AWS Bedrock (default)
- OpenAI / OpenAI-compatible APIs (via `OPENAI_BASE_URL`)
- Anthropic
- Google AI
- Azure OpenAI
- Ollama
Note that `claude-sonnet-4-5` has trained on draw.io diagrams with AWS logos, so if you want to create AWS architecture diagrams, this is the best choice.
- **LLM-Powered Diagram Creation**: Leverage Large Language Models to create and manipulate draw.io diagrams directly through natural language commands
- **Image-Based Diagram Replication**: Upload existing diagrams or images and have the AI replicate and enhance them automatically
- **PDF & Text File Upload**: Upload PDF documents and text files to extract content and generate diagrams from existing documents
- **AI Reasoning Display**: View the AI's thinking process for supported models (OpenAI o1/o3, Gemini, Claude, etc.)
- **Diagram History**: Comprehensive version control that tracks all changes, allowing you to view and restore previous versions of your diagrams before the AI editing.
- **Interactive Chat Interface**: Communicate with AI to refine your diagrams in real-time
- **Cloud Architecture Diagram Support**: Specialized support for generating cloud architecture diagrams (AWS, GCP, Azure)
- **Animated Connectors**: Create dynamic and animated connectors between diagram elements for better visualization
## Getting Started
### Try it Online
No installation needed! Try the app directly on our demo site:
[![Live Demo](./public/live-demo-button.svg)](https://next-ai-drawio.jiang.jp/)
> Note: Due to high traffic, the demo site currently uses minimax-m2. For best results, we recommend self-hosting with Claude Sonnet 4.5 or Claude Opus 4.5.
> **Bring Your Own API Key**: You can use your own API key to bypass usage limits on the demo site. Click the Settings icon in the chat panel to configure your provider and API key. Your key is stored locally in your browser and is never stored on the server.
### Run with Docker (Recommended)
If you just want to run it locally, the best way is to use Docker.
First, install Docker if you haven't already: [Get Docker](https://docs.docker.com/get-docker/)
Then run:
```bash
docker run -d -p 3000:3000 \
-e AI_PROVIDER=openai \
-e AI_MODEL=gpt-4o \
-e OPENAI_API_KEY=your_api_key \
ghcr.io/dayuanjiang/next-ai-draw-io:latest
```
Or use an env file:
```bash
cp env.example .env
# Edit .env with your configuration
docker run -d -p 3000:3000 --env-file .env ghcr.io/dayuanjiang/next-ai-draw-io:latest
```
Open [http://localhost:3000](http://localhost:3000) in your browser.
Replace the environment variables with your preferred AI provider configuration. See [Multi-Provider Support](#multi-provider-support) for available options.
> **Offline Deployment:** If `embed.diagrams.net` is blocked, see [Offline Deployment](./docs/offline-deployment.md) for configuration options.
### Installation
1. Clone the repository:
@@ -91,8 +147,6 @@ cd next-ai-draw-io
```bash
npm install
# or
yarn install
```
3. Configure your AI provider:
@@ -105,11 +159,15 @@ cp env.example .env.local
Edit `.env.local` and configure your chosen provider:
- Set `AI_PROVIDER` to your chosen provider (bedrock, openai, anthropic, google, azure, ollama)
- Set `AI_PROVIDER` to your chosen provider (bedrock, openai, anthropic, google, azure, ollama, openrouter, deepseek, siliconflow)
- Set `AI_MODEL` to the specific model you want to use
- Add the required API keys for your provider
- `TEMPERATURE`: Optional temperature setting (e.g., `0` for deterministic output). Leave unset for models that don't support it (e.g., reasoning models).
- `ACCESS_CODE_LIST`: Optional access password(s), can be comma-separated for multiple passwords.
See the [Multi-Provider Support](#multi-provider-support) section above for provider-specific configuration examples.
> Warning: If you do not set `ACCESS_CODE_LIST`, anyone can access your deployed site directly, which may lead to rapid depletion of your token. It is recommended to set this option.
See the [Provider Configuration Guide](./docs/ai-providers.md) for detailed setup instructions for each provider.
4. Run the development server:
@@ -130,35 +188,67 @@ Or you can deploy by this button.
Be sure to **set the environment variables** in the Vercel dashboard as you did in your local `.env.local` file.
## Multi-Provider Support
- AWS Bedrock (default)
- OpenAI
- Anthropic
- Google AI
- Azure OpenAI
- Ollama
- OpenRouter
- DeepSeek
- SiliconFlow
All providers except AWS Bedrock and OpenRouter support custom endpoints.
📖 **[Detailed Provider Configuration Guide](./docs/ai-providers.md)** - See setup instructions for each provider.
**Model Requirements**: This task requires strong model capabilities for generating long-form text with strict formatting constraints (draw.io XML). Recommended models include Claude Sonnet 4.5, GPT-5.1, Gemini 3 Pro, and DeepSeek V3.2/R1.
Note that `claude` series has trained on draw.io diagrams with cloud architecture logos like AWS, Azure, GCP. So if you want to create cloud architecture diagrams, this is the best choice.
## How It Works
The application uses the following technologies:
- **Next.js**: For the frontend framework and routing
- **Vercel AI SDK** (`ai` + `@ai-sdk/*`): For streaming AI responses and multi-provider support
- **react-drawio**: For diagram representation and manipulation
Diagrams are represented as XML that can be rendered in draw.io. The AI processes your commands and generates or modifies this XML accordingly.
## Project Structure
```
app/ # Next.js application routes and pages
extract_xml.ts # Utilities for XML processing
app/ # Next.js App Router
api/chat/ # Chat API endpoint with AI tools
page.tsx # Main page with DrawIO embed
components/ # React components
chat-input.tsx # User input component for AI interaction
chatPanel.tsx # Chat interface with diagram control
chat-panel.tsx # Chat interface with diagram control
chat-input.tsx # User input component with file upload
history-dialog.tsx # Diagram version history viewer
ui/ # UI components (buttons, cards, etc.)
contexts/ # React context providers
diagram-context.tsx # Global diagram state management
lib/ # Utility functions and helpers
utils.ts # General utilities including XML conversion
ai-providers.ts # Multi-provider AI configuration
utils.ts # XML processing and conversion utilities
public/ # Static assets including example images
```
## TODOs
- [x] Allow the LLM to modify the XML instead of generating it from scratch everytime.
- [x] Improve the smoothness of shape streaming updates.
- [x] Add multiple AI provider support (OpenAI, Anthropic, Google, Azure, Ollama)
- [ ] Solve the bug that generation will fail for session that longer than 60s.
## License
This project is licensed under the MIT License.
## Support & Contact
If you find this project useful, please consider [sponsoring](https://github.com/sponsors/DayuanJiang) to help me host the live demo site!
For support or inquiries, please open an issue on the GitHub repository or contact the maintainer at:
- Email: me[at]jiang.jp
## Star History
[![Star History Chart](https://api.star-history.com/svg?repos=DayuanJiang/next-ai-draw-io&type=date&legend=top-left)](https://www.star-history.com/#DayuanJiang/next-ai-draw-io&type=date&legend=top-left)
---

457
app/about/cn/page.tsx Normal file
View File

@@ -0,0 +1,457 @@
import type { Metadata } from "next"
import Image from "next/image"
import Link from "next/link"
import { FaGithub } from "react-icons/fa"
export const metadata: Metadata = {
title: "关于 - Next AI Draw.io",
description:
"AI驱动的图表创建工具 - 对话、绘制、可视化。使用自然语言创建AWS、GCP和Azure架构图。",
keywords: ["AI图表", "draw.io", "AWS架构", "GCP图表", "Azure图表", "LLM"],
}
function formatNumber(num: number): string {
if (num >= 1000) {
return `${num / 1000}k`
}
return num.toString()
}
export default function AboutCN() {
const dailyRequestLimit = Number(process.env.DAILY_REQUEST_LIMIT) || 20
const dailyTokenLimit = Number(process.env.DAILY_TOKEN_LIMIT) || 500000
const tpmLimit = Number(process.env.TPM_LIMIT) || 50000
return (
<div className="min-h-screen bg-gray-50">
{/* Navigation */}
<header className="bg-white border-b border-gray-200">
<div className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-4">
<div className="flex items-center justify-between">
<Link
href="/"
className="text-xl font-bold text-gray-900 hover:text-gray-700"
>
Next AI Draw.io
</Link>
<nav className="flex items-center gap-6 text-sm">
<Link
href="/"
className="text-gray-600 hover:text-gray-900 transition-colors"
>
</Link>
<Link
href="/about/cn"
className="text-blue-600 font-semibold"
>
</Link>
<a
href="https://github.com/DayuanJiang/next-ai-draw-io"
target="_blank"
rel="noopener noreferrer"
className="text-gray-600 hover:text-gray-900 transition-colors"
aria-label="在GitHub上查看"
>
<FaGithub className="w-5 h-5" />
</a>
</nav>
</div>
</div>
</header>
{/* Main Content */}
<main className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-12">
<article className="prose prose-lg max-w-none">
{/* Title */}
<div className="text-center mb-8">
<h1 className="text-4xl font-bold text-gray-900 mb-2">
Next AI Draw.io
</h1>
<p className="text-xl text-gray-600 font-medium">
AI驱动的图表创建工具 -
</p>
<div className="flex justify-center gap-4 mt-4 text-sm">
<Link
href="/about"
className="text-gray-600 hover:text-blue-600"
>
English
</Link>
<span className="text-gray-400">|</span>
<Link
href="/about/cn"
className="text-blue-600 font-semibold"
>
</Link>
<span className="text-gray-400">|</span>
<Link
href="/about/ja"
className="text-gray-600 hover:text-blue-600"
>
</Link>
</div>
</div>
<div className="relative mb-8 rounded-2xl bg-gradient-to-br from-amber-50 via-orange-50 to-rose-50 p-[1px] shadow-lg">
<div className="absolute inset-0 rounded-2xl bg-gradient-to-br from-amber-400 via-orange-400 to-rose-400 opacity-20" />
<div className="relative rounded-2xl bg-white/80 backdrop-blur-sm p-6">
{/* Header */}
<div className="mb-4">
<h3 className="text-lg font-bold text-gray-900 tracking-tight">
{" "}
<span className="text-sm text-amber-600 font-medium italic font-normal">
()
</span>
</h3>
</div>
{/* Story */}
<div className="space-y-3 text-sm text-gray-700 leading-relaxed mb-5">
<p>
AI
(TPS/TPM)
</p>
<p>
使 Claude {" "}
<span className="font-semibold text-amber-700">
minimax-m2
</span>
</p>
<p>
<span className="font-semibold text-amber-700">
</span>
API
</p>
</div>
{/* Limits Cards */}
<div className="grid grid-cols-2 gap-3 mb-5">
<div className="rounded-xl bg-gradient-to-br from-amber-100 to-orange-100 p-4 text-center">
<div className="text-xs font-medium text-amber-700 uppercase tracking-wide mb-1">
Token
</div>
<div className="text-lg font-bold text-gray-900">
{formatNumber(tpmLimit)}
<span className="text-sm font-normal text-gray-600">
/
</span>
</div>
<div className="text-lg font-bold text-gray-900">
{formatNumber(dailyTokenLimit)}
<span className="text-sm font-normal text-gray-600">
/
</span>
</div>
</div>
<div className="rounded-xl bg-gradient-to-br from-amber-100 to-orange-100 p-4 text-center">
<div className="text-xs font-medium text-amber-700 uppercase tracking-wide mb-1">
</div>
<div className="text-2xl font-bold text-gray-900">
{dailyRequestLimit}
</div>
<div className="text-sm text-gray-600">
</div>
</div>
</div>
{/* Divider */}
<div className="flex items-center gap-3 my-5">
<div className="flex-1 h-px bg-gradient-to-r from-transparent via-amber-300 to-transparent" />
</div>
{/* Bring Your Own Key */}
<div className="text-center mb-5">
<h4 className="text-base font-bold text-gray-900 mb-2">
使 API Key
</h4>
<p className="text-sm text-gray-600 mb-2 max-w-md mx-auto">
使 API Key
Provider API Key
</p>
<p className="text-xs text-gray-500 max-w-md mx-auto">
Key
</p>
</div>
{/* Divider */}
<div className="flex items-center gap-3 mb-5">
<div className="flex-1 h-px bg-gradient-to-r from-transparent via-amber-300 to-transparent" />
</div>
{/* Sponsorship CTA */}
<div className="text-center">
<h4 className="text-base font-bold text-gray-900 mb-2">
()
</h4>
<p className="text-sm text-gray-600 mb-4 max-w-md mx-auto">
AI API
</p>
<p className="text-sm text-gray-600 mb-4 max-w-md mx-auto">
GitHub Live Demo
Logo
</p>
<a
href="mailto:me@jiang.jp"
className="inline-flex items-center gap-2 px-5 py-2.5 rounded-full bg-gradient-to-r from-amber-500 to-orange-500 text-white font-medium text-sm shadow-md hover:shadow-lg hover:scale-105 transition-all duration-200"
>
</a>
</div>
</div>
</div>
<p className="text-gray-700">
AI功能的Next.js网页应用draw.io图表无缝结合AI辅助可视化来创建
</p>
{/* Features */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<ul className="list-disc pl-6 text-gray-700 space-y-2">
<li>
<strong>LLM驱动的图表创建</strong>
draw.io图表
</li>
<li>
<strong></strong>
AI自动复制和增强
</li>
<li>
<strong></strong>
AI编辑前的图表版本
</li>
<li>
<strong></strong>
AI实时对话来完善您的图表
</li>
<li>
<strong>AWS架构图支持</strong>
AWS架构图
</li>
<li>
<strong></strong>
</li>
</ul>
{/* Examples */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<p className="text-gray-700 mb-6">
</p>
<div className="space-y-8">
{/* Animated Transformer */}
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
Transformer连接器
</h3>
<p className="text-gray-600 mb-4">
<strong></strong>
<strong></strong>Transformer架构图
</p>
<Image
src="/animated_connectors.svg"
alt="带动画连接器的Transformer架构"
width={480}
height={360}
className="mx-auto"
/>
</div>
{/* Cloud Architecture Grid */}
<div className="grid md:grid-cols-2 gap-6">
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
GCP架构图
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong> 使
<strong>GCP图标</strong>
GCP架构图
</p>
<Image
src="/gcp_demo.svg"
alt="GCP架构图"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
AWS架构图
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong> 使
<strong>AWS图标</strong>
AWS架构图
</p>
<Image
src="/aws_demo.svg"
alt="AWS架构图"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
Azure架构图
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong> 使
<strong>Azure图标</strong>
Azure架构图
</p>
<Image
src="/azure_demo.svg"
alt="Azure架构图"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong>{" "}
</p>
<Image
src="/cat_demo.svg"
alt="猫咪绘图"
width={240}
height={240}
className="mx-auto"
/>
</div>
</div>
</div>
{/* How It Works */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<p className="text-gray-700 mb-4">使</p>
<ul className="list-disc pl-6 text-gray-700 space-y-2">
<li>
<strong>Next.js</strong>
</li>
<li>
<strong>Vercel AI SDK</strong><code>ai</code> +{" "}
<code>@ai-sdk/*</code>
用于流式AI响应和多提供商支持
</li>
<li>
<strong>react-drawio</strong>:用于图表表示和操作
</li>
</ul>
<p className="text-gray-700 mt-4">
图表以XML格式表示可在draw.io中渲染。AI处理您的命令并相应地生成或修改此XML。
</p>
{/* Multi-Provider Support */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<ul className="list-disc pl-6 text-gray-700 space-y-1">
<li>AWS Bedrock</li>
<li>
OpenAI / OpenAI兼容API{" "}
<code>OPENAI_BASE_URL</code>
</li>
<li>Anthropic</li>
<li>Google AI</li>
<li>Azure OpenAI</li>
<li>Ollama</li>
<li>OpenRouter</li>
<li>DeepSeek</li>
</ul>
<p className="text-gray-700 mt-4">
<code>claude-sonnet-4-5</code>{" "}
AWS标志的draw.io图表上进行训练AWS架构图
</p>
{/* Support */}
<div className="flex items-center gap-4 mt-10 mb-4">
<h2 className="text-2xl font-semibold text-gray-900">
</h2>
<iframe
src="https://github.com/sponsors/DayuanJiang/button"
title="Sponsor DayuanJiang"
height="32"
width="114"
style={{ border: 0, borderRadius: 6 }}
/>
</div>
<p className="text-gray-700">
{" "}
<a
href="https://github.com/sponsors/DayuanJiang"
target="_blank"
rel="noopener noreferrer"
className="text-blue-600 hover:underline"
>
</a>{" "}
线
</p>
<p className="text-gray-700 mt-2">
{" "}
<a
href="https://github.com/DayuanJiang/next-ai-draw-io"
target="_blank"
rel="noopener noreferrer"
className="text-blue-600 hover:underline"
>
GitHub仓库
</a>{" "}
issue或联系me[at]jiang.jp
</p>
{/* CTA */}
<div className="mt-12 text-center">
<Link
href="/"
className="inline-block bg-blue-600 text-white px-8 py-3 rounded-lg font-semibold hover:bg-blue-700 transition-colors"
>
</Link>
</div>
</article>
</main>
{/* Footer */}
<footer className="bg-white border-t border-gray-200 mt-16">
<div className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-6">
<p className="text-center text-gray-600 text-sm">
Next AI Draw.io - AI驱动的图表生成器
</p>
</div>
</footer>
</div>
)
}

472
app/about/ja/page.tsx Normal file
View File

@@ -0,0 +1,472 @@
import type { Metadata } from "next"
import Image from "next/image"
import Link from "next/link"
import { FaGithub } from "react-icons/fa"
export const metadata: Metadata = {
title: "概要 - Next AI Draw.io",
description:
"AI搭載のダイアグラム作成ツール - チャット、描画、可視化。自然言語でAWS、GCP、Azureアーキテクチャ図を作成。",
keywords: [
"AIダイアグラム",
"draw.io",
"AWSアーキテクチャ",
"GCPダイアグラム",
"Azureダイアグラム",
"LLM",
],
}
function formatNumber(num: number): string {
if (num >= 1000) {
return `${num / 1000}k`
}
return num.toString()
}
export default function AboutJA() {
const dailyRequestLimit = Number(process.env.DAILY_REQUEST_LIMIT) || 20
const dailyTokenLimit = Number(process.env.DAILY_TOKEN_LIMIT) || 500000
const tpmLimit = Number(process.env.TPM_LIMIT) || 50000
return (
<div className="min-h-screen bg-gray-50">
{/* Navigation */}
<header className="bg-white border-b border-gray-200">
<div className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-4">
<div className="flex items-center justify-between">
<Link
href="/"
className="text-xl font-bold text-gray-900 hover:text-gray-700"
>
Next AI Draw.io
</Link>
<nav className="flex items-center gap-6 text-sm">
<Link
href="/"
className="text-gray-600 hover:text-gray-900 transition-colors"
>
</Link>
<Link
href="/about/ja"
className="text-blue-600 font-semibold"
>
</Link>
<a
href="https://github.com/DayuanJiang/next-ai-draw-io"
target="_blank"
rel="noopener noreferrer"
className="text-gray-600 hover:text-gray-900 transition-colors"
aria-label="GitHubで見る"
>
<FaGithub className="w-5 h-5" />
</a>
</nav>
</div>
</div>
</header>
{/* Main Content */}
<main className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-12">
<article className="prose prose-lg max-w-none">
{/* Title */}
<div className="text-center mb-8">
<h1 className="text-4xl font-bold text-gray-900 mb-2">
Next AI Draw.io
</h1>
<p className="text-xl text-gray-600 font-medium">
AI搭載のダイアグラム作成ツール -
</p>
<div className="flex justify-center gap-4 mt-4 text-sm">
<Link
href="/about"
className="text-gray-600 hover:text-blue-600"
>
English
</Link>
<span className="text-gray-400">|</span>
<Link
href="/about/cn"
className="text-gray-600 hover:text-blue-600"
>
</Link>
<span className="text-gray-400">|</span>
<Link
href="/about/ja"
className="text-blue-600 font-semibold"
>
</Link>
</div>
</div>
<div className="relative mb-8 rounded-2xl bg-gradient-to-br from-amber-50 via-orange-50 to-rose-50 p-[1px] shadow-lg">
<div className="absolute inset-0 rounded-2xl bg-gradient-to-br from-amber-400 via-orange-400 to-rose-400 opacity-20" />
<div className="relative rounded-2xl bg-white/80 backdrop-blur-sm p-6">
{/* Header */}
<div className="mb-4">
<h3 className="text-lg font-bold text-gray-900 tracking-tight">
{" "}
<span className="text-sm text-amber-600 font-medium italic font-normal">
</span>
</h3>
</div>
{/* Story */}
<div className="space-y-3 text-sm text-gray-700 leading-relaxed mb-5">
<p>
AI API (TPS/TPM)
</p>
<p>
Claude {" "}
<span className="font-semibold text-amber-700">
minimax-m2
</span>{" "}
</p>
<p>
<span className="font-semibold text-amber-700">
</span>
API
</p>
</div>
{/* Limits Cards */}
<div className="grid grid-cols-2 gap-3 mb-5">
<div className="rounded-xl bg-gradient-to-br from-amber-100 to-orange-100 p-4 text-center">
<div className="text-xs font-medium text-amber-700 uppercase tracking-wide mb-1">
使
</div>
<div className="text-lg font-bold text-gray-900">
{formatNumber(tpmLimit)}
<span className="text-sm font-normal text-gray-600">
/
</span>
</div>
<div className="text-lg font-bold text-gray-900">
{formatNumber(dailyTokenLimit)}
<span className="text-sm font-normal text-gray-600">
/
</span>
</div>
</div>
<div className="rounded-xl bg-gradient-to-br from-amber-100 to-orange-100 p-4 text-center">
<div className="text-xs font-medium text-amber-700 uppercase tracking-wide mb-1">
1
</div>
<div className="text-2xl font-bold text-gray-900">
{dailyRequestLimit}
</div>
<div className="text-sm text-gray-600">
</div>
</div>
</div>
{/* Divider */}
<div className="flex items-center gap-3 my-5">
<div className="flex-1 h-px bg-gradient-to-r from-transparent via-amber-300 to-transparent" />
</div>
{/* Bring Your Own Key */}
<div className="text-center mb-5">
<h4 className="text-base font-bold text-gray-900 mb-2">
APIキーを使用
</h4>
<p className="text-sm text-gray-600 mb-2 max-w-md mx-auto">
APIキーを使用することでAPIキーを設定してください
</p>
<p className="text-xs text-gray-500 max-w-md mx-auto">
</p>
</div>
{/* Divider */}
<div className="flex items-center gap-3 mb-5">
<div className="flex-1 h-px bg-gradient-to-r from-transparent via-amber-300 to-transparent" />
</div>
{/* Sponsorship CTA */}
<div className="text-center">
<h4 className="text-base font-bold text-gray-900 mb-2">
</h4>
<p className="text-sm text-gray-600 mb-4 max-w-md mx-auto">
AI
API
</p>
<p className="text-sm text-gray-600 mb-4 max-w-md mx-auto">
GitHub
</p>
<a
href="mailto:me@jiang.jp"
className="inline-flex items-center gap-2 px-5 py-2.5 rounded-full bg-gradient-to-r from-amber-500 to-orange-500 text-white font-medium text-sm shadow-md hover:shadow-lg hover:scale-105 transition-all duration-200"
>
</a>
</div>
</div>
</div>
<p className="text-gray-700">
AI機能とdraw.ioダイアグラムを統合したNext.jsウェブアプリケーションですAI支援の可視化により
</p>
{/* Features */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<ul className="list-disc pl-6 text-gray-700 space-y-2">
<li>
<strong>LLM搭載のダイアグラム作成</strong>
draw.ioダイアグラムを作成
</li>
<li>
<strong></strong>
AIが自動的に複製
</li>
<li>
<strong></strong>
AI編集前のダイアグラムの以前のバージョンを表示
</li>
<li>
<strong>
</strong>
AIとリアルタイムでコミュニケーションしてダイアグラムを改善
</li>
<li>
<strong>
AWSアーキテクチャダイアグラムサポート
</strong>
AWSアーキテクチャダイアグラムの生成を専門的にサポート
</li>
<li>
<strong></strong>
</li>
</ul>
{/* Examples */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<p className="text-gray-700 mb-6">
</p>
<div className="space-y-8">
{/* Animated Transformer */}
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
Transformerコネクタ
</h3>
<p className="text-gray-600 mb-4">
<strong></strong>{" "}
<strong></strong>
Transformerアーキテクチャ図を作成してください
</p>
<Image
src="/animated_connectors.svg"
alt="アニメーションコネクタ付きTransformerアーキテクチャ"
width={480}
height={360}
className="mx-auto"
/>
</div>
{/* Cloud Architecture Grid */}
<div className="grid md:grid-cols-2 gap-6">
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
GCPアーキテクチャ図
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong>{" "}
<strong>GCPアイコン</strong>
使GCPアーキテクチャ図を生成してください
</p>
<Image
src="/gcp_demo.svg"
alt="GCPアーキテクチャ図"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
AWSアーキテクチャ図
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong>{" "}
<strong>AWSアイコン</strong>
使AWSアーキテクチャ図を生成してください
</p>
<Image
src="/aws_demo.svg"
alt="AWSアーキテクチャ図"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
Azureアーキテクチャ図
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong>{" "}
<strong>Azureアイコン</strong>
使Azureアーキテクチャ図を生成してください
</p>
<Image
src="/azure_demo.svg"
alt="Azureアーキテクチャ図"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
</h3>
<p className="text-gray-600 text-sm mb-4">
<strong></strong>{" "}
</p>
<Image
src="/cat_demo.svg"
alt="猫の絵"
width={240}
height={240}
className="mx-auto"
/>
</div>
</div>
</div>
{/* How It Works */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<p className="text-gray-700 mb-4">
使
</p>
<ul className="list-disc pl-6 text-gray-700 space-y-2">
<li>
<strong>Next.js</strong>
</li>
<li>
<strong>Vercel AI SDK</strong><code>ai</code> +{" "}
<code>@ai-sdk/*</code>
ストリーミングAIレスポンスとマルチプロバイダーサポート
</li>
<li>
<strong>react-drawio</strong>
:ダイアグラムの表現と操作
</li>
</ul>
<p className="text-gray-700 mt-4">
ダイアグラムはdraw.ioでレンダリングできるXMLとして表現されます。AIがコマンドを処理し、それに応じてこのXMLを生成または変更します。
</p>
{/* Multi-Provider Support */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
</h2>
<ul className="list-disc pl-6 text-gray-700 space-y-1">
<li>AWS Bedrock</li>
<li>
OpenAI / OpenAI互換API<code>OPENAI_BASE_URL</code>
</li>
<li>Anthropic</li>
<li>Google AI</li>
<li>Azure OpenAI</li>
<li>Ollama</li>
<li>OpenRouter</li>
<li>DeepSeek</li>
</ul>
<p className="text-gray-700 mt-4">
<code>claude-sonnet-4-5</code>
AWSロゴ付きのdraw.ioダイアグラムで学習されているためAWSアーキテクチャダイアグラムを作成したい場合は最適な選択です
</p>
{/* Support */}
<div className="flex items-center gap-4 mt-10 mb-4">
<h2 className="text-2xl font-semibold text-gray-900">
</h2>
<iframe
src="https://github.com/sponsors/DayuanJiang/button"
title="Sponsor DayuanJiang"
height="32"
width="114"
style={{ border: 0, borderRadius: 6 }}
/>
</div>
<p className="text-gray-700">
{" "}
<a
href="https://github.com/sponsors/DayuanJiang"
target="_blank"
rel="noopener noreferrer"
className="text-blue-600 hover:underline"
>
</a>{" "}
</p>
<p className="text-gray-700 mt-2">
{" "}
<a
href="https://github.com/DayuanJiang/next-ai-draw-io"
target="_blank"
rel="noopener noreferrer"
className="text-blue-600 hover:underline"
>
GitHubリポジトリ
</a>{" "}
issueを開くかme[at]jiang.jp
</p>
{/* CTA */}
<div className="mt-12 text-center">
<Link
href="/"
className="inline-block bg-blue-600 text-white px-8 py-3 rounded-lg font-semibold hover:bg-blue-700 transition-colors"
>
</Link>
</div>
</article>
</main>
{/* Footer */}
<footer className="bg-white border-t border-gray-200 mt-16">
<div className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-6">
<p className="text-center text-gray-600 text-sm">
Next AI Draw.io -
AI搭載ダイアグラムジェネレーター
</p>
</div>
</footer>
</div>
)
}

View File

@@ -1,28 +1,57 @@
import type { Metadata } from "next";
import Link from "next/link";
import { FaGithub } from "react-icons/fa";
import type { Metadata } from "next"
import Image from "next/image"
import Link from "next/link"
import { FaGithub } from "react-icons/fa"
export const metadata: Metadata = {
title: "About - AI-Powered Diagram Generator | Next AI Draw.io",
description: "Learn about Next AI Draw.io, a free AI-powered diagram creation tool. Create AWS architecture diagrams, flowcharts, and UML diagrams using Claude Sonnet and GPT-4. No login required.",
keywords: ["about AI diagram generator", "diagram tool features", "how to create diagrams", "AI drawing tool capabilities", "draw.io integration"],
};
title: "About - Next AI Draw.io",
description:
"AI-Powered Diagram Creation Tool - Chat, Draw, Visualize. Create AWS, GCP, and Azure architecture diagrams with natural language.",
keywords: [
"AI diagram",
"draw.io",
"AWS architecture",
"GCP diagram",
"Azure diagram",
"LLM",
],
}
function formatNumber(num: number): string {
if (num >= 1000) {
return `${num / 1000}k`
}
return num.toString()
}
export default function About() {
const dailyRequestLimit = Number(process.env.DAILY_REQUEST_LIMIT) || 20
const dailyTokenLimit = Number(process.env.DAILY_TOKEN_LIMIT) || 500000
const tpmLimit = Number(process.env.TPM_LIMIT) || 50000
return (
<div className="min-h-screen bg-gray-50">
{/* Navigation */}
<header className="bg-white border-b border-gray-200">
<div className="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 py-4">
<div className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-4">
<div className="flex items-center justify-between">
<Link href="/" className="text-xl font-bold text-gray-900 hover:text-gray-700">
<Link
href="/"
className="text-xl font-bold text-gray-900 hover:text-gray-700"
>
Next AI Draw.io
</Link>
<nav className="flex items-center gap-6 text-sm">
<Link href="/" className="text-gray-600 hover:text-gray-900 transition-colors">
<Link
href="/"
className="text-gray-600 hover:text-gray-900 transition-colors"
>
Editor
</Link>
<Link href="/about" className="text-blue-600 font-semibold">
<Link
href="/about"
className="text-blue-600 font-semibold"
>
About
</Link>
<a
@@ -40,297 +69,432 @@ export default function About() {
</header>
{/* Main Content */}
<main className="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 py-12">
<article>
{/* Hero Section */}
<header className="mb-12">
<h1 className="text-4xl font-bold text-gray-900 mb-4">
AI-Powered Diagram Generator | Create Professional Diagrams Instantly
<main className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-12">
<article className="prose prose-lg max-w-none">
{/* Title */}
<div className="text-center mb-8">
<h1 className="text-4xl font-bold text-gray-900 mb-2">
Next AI Draw.io
</h1>
<p className="text-xl text-gray-600">
Free, open-source diagram creation tool powered by AI. No login required, no installation needed.
<p className="text-xl text-gray-600 font-medium">
AI-Powered Diagram Creation Tool - Chat, Draw,
Visualize
</p>
</header>
{/* Introduction */}
<section className="mb-12">
<h2 className="text-2xl font-semibold text-gray-900 mb-4">What is Next AI Draw.io?</h2>
<div className="prose prose-lg max-w-none text-gray-700">
<p className="mb-4">
<strong>Next AI Draw.io</strong> is a free, AI-powered diagram creation tool that integrates seamlessly with draw.io.
Generate AWS architecture diagrams, flowcharts, UML diagrams, and technical documentation diagrams using natural language
prompts. No login required, no installation neededstart creating professional diagrams instantly in your browser.
</p>
<p className="mb-4">
Our intelligent diagram generator uses advanced AI models including <strong>Claude Sonnet</strong> and <strong>GPT-4</strong> to
understand your requirements and automatically create properly structured diagrams with appropriate symbols, layouts, and connections.
Simply describe what you need, upload reference images, or ask the AI to modify existing diagrams with our targeted XML editing feature.
</p>
<p>
Whether you're a software architect designing system infrastructure, a developer documenting APIs, a business analyst creating
process flows, or a student working on technical assignments, Next AI Draw.io makes diagram creation fast, accurate, and effortless.
</p>
<div className="flex justify-center gap-4 mt-4 text-sm">
<Link
href="/about"
className="text-blue-600 font-semibold"
>
English
</Link>
<span className="text-gray-400">|</span>
<Link
href="/about/cn"
className="text-gray-600 hover:text-blue-600"
>
</Link>
<span className="text-gray-400">|</span>
<Link
href="/about/ja"
className="text-gray-600 hover:text-blue-600"
>
</Link>
</div>
</section>
</div>
{/* Key Features */}
<section className="mb-12">
<h2 className="text-2xl font-semibold text-gray-900 mb-6">Key Features</h2>
<div className="relative mb-8 rounded-2xl bg-gradient-to-br from-amber-50 via-orange-50 to-rose-50 p-[1px] shadow-lg">
<div className="absolute inset-0 rounded-2xl bg-gradient-to-br from-amber-400 via-orange-400 to-rose-400 opacity-20" />
<div className="relative rounded-2xl bg-white/80 backdrop-blur-sm p-6">
{/* Header */}
<div className="mb-4">
<h3 className="text-lg font-bold text-gray-900 tracking-tight">
Model Change & Usage Limits{" "}
<span className="text-sm text-amber-600 font-medium italic font-normal">
(Or: Why My Wallet is Crying)
</span>
</h3>
</div>
{/* Story */}
<div className="space-y-3 text-sm text-gray-700 leading-relaxed mb-5">
<p>
The response to this project has been
incredibleyou all love making diagrams!
However, this enthusiasm means we are
frequently hitting the AI API rate limits
(TPS/TPM). When this happens, the system
pauses, leading to failed requests.
</p>
<p>
Due to the high usage, I have changed the
model from Claude to{" "}
<span className="font-semibold text-amber-700">
minimax-m2
</span>
, which is more cost-effective.
</p>
<p>
As an{" "}
<span className="font-semibold text-amber-700">
indie developer
</span>
, I am currently footing the entire API
bill. To keep the lights on and ensure the
service remains available to everyone
without sending me into debt, I have also
implemented the following temporary caps:
</p>
</div>
{/* Limits Cards */}
<div className="grid grid-cols-2 gap-3 mb-5">
<div className="rounded-xl bg-gradient-to-br from-amber-100 to-orange-100 p-4 text-center">
<div className="text-xs font-medium text-amber-700 uppercase tracking-wide mb-1">
Token Usage
</div>
<div className="text-lg font-bold text-gray-900">
{formatNumber(tpmLimit)}
<span className="text-sm font-normal text-gray-600">
/min
</span>
</div>
<div className="text-lg font-bold text-gray-900">
{formatNumber(dailyTokenLimit)}
<span className="text-sm font-normal text-gray-600">
/day
</span>
</div>
</div>
<div className="rounded-xl bg-gradient-to-br from-amber-100 to-orange-100 p-4 text-center">
<div className="text-xs font-medium text-amber-700 uppercase tracking-wide mb-1">
Daily Requests
</div>
<div className="text-2xl font-bold text-gray-900">
{dailyRequestLimit}
</div>
<div className="text-sm text-gray-600">
requests
</div>
</div>
</div>
{/* Divider */}
<div className="flex items-center gap-3 my-5">
<div className="flex-1 h-px bg-gradient-to-r from-transparent via-amber-300 to-transparent" />
</div>
{/* Bring Your Own Key */}
<div className="text-center mb-5">
<h4 className="text-base font-bold text-gray-900 mb-2">
Bring Your Own API Key
</h4>
<p className="text-sm text-gray-600 mb-2 max-w-md mx-auto">
You can use your own API key to bypass these
limits. Click the Settings icon in the chat
panel to configure your provider and API
key.
</p>
<p className="text-xs text-gray-500 max-w-md mx-auto">
Your key is stored locally in your browser
and is never stored on the server.
</p>
</div>
{/* Divider */}
<div className="flex items-center gap-3 mb-5">
<div className="flex-1 h-px bg-gradient-to-r from-transparent via-amber-300 to-transparent" />
</div>
{/* Sponsorship CTA */}
<div className="text-center">
<h4 className="text-base font-bold text-gray-900 mb-2">
Call for Sponsorship
</h4>
<p className="text-sm text-gray-600 mb-4 max-w-md mx-auto">
Scaling the backend is the only way to
remove these limits. I am actively seeking
sponsorship from AI API providers or Cloud
Platforms.
</p>
<p className="text-sm text-gray-600 mb-4 max-w-md mx-auto">
In return for support (credits or funding),
I will prominently feature your company as a
platform sponsor on both the GitHub
repository and the live demo site.
</p>
<a
href="mailto:me@jiang.jp"
className="inline-flex items-center gap-2 px-5 py-2.5 rounded-full bg-gradient-to-r from-amber-500 to-orange-500 text-white font-medium text-sm shadow-md hover:shadow-lg hover:scale-105 transition-all duration-200"
>
Contact Me
</a>
</div>
</div>
</div>
<p className="text-gray-700">
A Next.js web application that integrates AI
capabilities with draw.io diagrams. Create, modify, and
enhance diagrams through natural language commands and
AI-assisted visualization.
</p>
{/* Features */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
Features
</h2>
<ul className="list-disc pl-6 text-gray-700 space-y-2">
<li>
<strong>LLM-Powered Diagram Creation</strong>:
Leverage Large Language Models to create and
manipulate draw.io diagrams directly through natural
language commands
</li>
<li>
<strong>Image-Based Diagram Replication</strong>:
Upload existing diagrams or images and have the AI
replicate and enhance them automatically
</li>
<li>
<strong>Diagram History</strong>: Comprehensive
version control that tracks all changes, allowing
you to view and restore previous versions of your
diagrams before the AI editing
</li>
<li>
<strong>Interactive Chat Interface</strong>:
Communicate with AI to refine your diagrams in
real-time
</li>
<li>
<strong>AWS Architecture Diagram Support</strong>:
Specialized support for generating AWS architecture
diagrams
</li>
<li>
<strong>Animated Connectors</strong>: Create dynamic
and animated connectors between diagram elements for
better visualization
</li>
</ul>
{/* Examples */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
Examples
</h2>
<p className="text-gray-700 mb-6">
Here are some example prompts and their generated
diagrams:
</p>
<div className="space-y-8">
{/* Animated Transformer */}
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
Animated Transformer Connectors
</h3>
<p className="text-gray-600 mb-4">
<strong>Prompt:</strong> Give me an{" "}
<strong>animated connector</strong> diagram of
transformer&apos;s architecture.
</p>
<Image
src="/animated_connectors.svg"
alt="Transformer Architecture with Animated Connectors"
width={480}
height={360}
className="mx-auto"
/>
</div>
{/* Cloud Architecture Grid */}
<div className="grid md:grid-cols-2 gap-6">
<div className="bg-white p-6 rounded-lg border border-gray-200 shadow-sm">
<h3 className="text-lg font-semibold text-gray-900 mb-3 flex items-center">
<span className="text-blue-600 mr-2">✓</span>
AI-Powered Diagram Creation
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
GCP Architecture Diagram
</h3>
<p className="text-gray-700">
Generate diagrams from natural language descriptions using Claude Sonnet or GPT-4.
Describe your diagram in plain English, and watch the AI create it with proper symbols,
layouts, and connections automatically.
<p className="text-gray-600 text-sm mb-4">
<strong>Prompt:</strong> Generate a GCP
architecture diagram with{" "}
<strong>GCP icons</strong>. Users connect to
a frontend hosted on an instance.
</p>
<Image
src="/gcp_demo.svg"
alt="GCP Architecture Diagram"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="bg-white p-6 rounded-lg border border-gray-200 shadow-sm">
<h3 className="text-lg font-semibold text-gray-900 mb-3 flex items-center">
<span className="text-blue-600 mr-2">✓</span>
AWS Architecture Diagrams
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
AWS Architecture Diagram
</h3>
<p className="text-gray-700">
Create professional cloud infrastructure diagrams with AWS-style icons and layouts.
Perfect for designing EC2 instances, Lambda functions, S3 buckets, RDS databases, VPCs,
and complete AWS solution architectures.
<p className="text-gray-600 text-sm mb-4">
<strong>Prompt:</strong> Generate an AWS
architecture diagram with{" "}
<strong>AWS icons</strong>. Users connect to
a frontend hosted on an instance.
</p>
<Image
src="/aws_demo.svg"
alt="AWS Architecture Diagram"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="bg-white p-6 rounded-lg border border-gray-200 shadow-sm">
<h3 className="text-lg font-semibold text-gray-900 mb-3 flex items-center">
<span className="text-blue-600 mr-2">✓</span>
Image-Based Diagram Replication
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
Azure Architecture Diagram
</h3>
<p className="text-gray-700">
Upload existing diagrams or sketches, and the AI will automatically recreate them in draw.io format.
Modify uploaded images by describing the changes you want—the AI handles the rest.
<p className="text-gray-600 text-sm mb-4">
<strong>Prompt:</strong> Generate an Azure
architecture diagram with{" "}
<strong>Azure icons</strong>. Users connect
to a frontend hosted on an instance.
</p>
<Image
src="/azure_demo.svg"
alt="Azure Architecture Diagram"
width={400}
height={300}
className="mx-auto"
/>
</div>
<div className="bg-white p-6 rounded-lg border border-gray-200 shadow-sm">
<h3 className="text-lg font-semibold text-gray-900 mb-3 flex items-center">
<span className="text-blue-600 mr-2">✓</span>
Diagram History & Version Control
<div className="text-center">
<h3 className="text-lg font-semibold text-gray-900 mb-2">
Cat Sketch
</h3>
<p className="text-gray-700">
Access previous versions of your diagrams and restore any version from your session history.
Never lose work—every AI modification is saved and can be undone with a single click.
</p>
</div>
<div className="bg-white p-6 rounded-lg border border-gray-200 shadow-sm">
<h3 className="text-lg font-semibold text-gray-900 mb-3 flex items-center">
<span className="text-blue-600 mr-2">✓</span>
Targeted XML Editing
</h3>
<p className="text-gray-700">
Precise diagram modifications using intelligent XML manipulation. Unlike full diagram regeneration,
targeted edits preserve your existing layout while making specific changes, ensuring consistent
and predictable results.
</p>
</div>
<div className="bg-white p-6 rounded-lg border border-gray-200 shadow-sm">
<h3 className="text-lg font-semibold text-gray-900 mb-3 flex items-center">
<span className="text-blue-600 mr-2">✓</span>
Multi-Provider AI Support
</h3>
<p className="text-gray-700">
Choose between Claude Sonnet, GPT-4, and other leading AI models for optimal results.
Each model has unique strengths—select the one that best fits your diagram complexity and style.
<p className="text-gray-600 text-sm mb-4">
<strong>Prompt:</strong> Draw a cute cat for
me.
</p>
<Image
src="/cat_demo.svg"
alt="Cat Drawing"
width={240}
height={240}
className="mx-auto"
/>
</div>
</div>
</section>
{/* Use Cases */}
<section className="mb-12">
<h2 className="text-2xl font-semibold text-gray-900 mb-6">Popular Use Cases</h2>
<div className="grid md:grid-cols-3 gap-6">
<div className="bg-blue-50 p-6 rounded-lg border border-blue-200">
<h3 className="text-xl font-semibold text-gray-900 mb-3">AWS Cloud Architecture</h3>
<p className="text-gray-700 mb-4">
Design scalable cloud infrastructure with EC2 instances, Lambda functions, S3 storage,
RDS databases, and VPC networking. Perfect for solution architects, cloud engineers,
and DevOps teams planning AWS deployments.
</p>
<p className="text-sm text-gray-600 italic">
Example: "Create an AWS diagram with an Application Load Balancer, two EC2 instances
in different availability zones, an RDS database, and an S3 bucket for static assets."
</p>
</div>
<div className="bg-green-50 p-6 rounded-lg border border-green-200">
<h3 className="text-xl font-semibold text-gray-900 mb-3">Flowcharts & Process Diagrams</h3>
<p className="text-gray-700 mb-4">
Create business process flows, decision trees, workflow diagrams, and algorithm flowcharts
for documentation, presentations, and process optimization. Ideal for business analysts,
project managers, and operations teams.
</p>
<p className="text-sm text-gray-600 italic">
Example: "Draw a flowchart for user authentication: check if user exists, verify password,
generate JWT token on success, show error message on failure."
</p>
</div>
<div className="bg-purple-50 p-6 rounded-lg border border-purple-200">
<h3 className="text-xl font-semibold text-gray-900 mb-3">System Design & UML Diagrams</h3>
<p className="text-gray-700 mb-4">
Generate system architecture diagrams, class diagrams, sequence diagrams, and
entity-relationship diagrams for software projects. Essential for software engineers,
system designers, and technical documentation.
</p>
<p className="text-sm text-gray-600 italic">
Example: "Create a class diagram for an e-commerce system with User, Product, Order,
and Payment classes showing their relationships and key methods."
</p>
</div>
</div>
</section>
</div>
{/* How It Works */}
<section className="mb-12 bg-white p-8 rounded-lg border border-gray-200">
<h2 className="text-2xl font-semibold text-gray-900 mb-6">How to Use Next AI Draw.io</h2>
<div className="space-y-6">
<div className="flex items-start">
<div className="flex-shrink-0 w-10 h-10 bg-blue-600 text-white rounded-full flex items-center justify-center font-bold mr-4">
1
</div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">Open the Editor</h3>
<p className="text-gray-700">
Navigate to the main page and you'll see the draw.io editor with an AI chat panel on the right.
No account creation or login requiredstart immediately.
</p>
</div>
</div>
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
How It Works
</h2>
<p className="text-gray-700 mb-4">
The application uses the following technologies:
</p>
<ul className="list-disc pl-6 text-gray-700 space-y-2">
<li>
<strong>Next.js</strong>: For the frontend framework
and routing
</li>
<li>
<strong>Vercel AI SDK</strong> (<code>ai</code> +{" "}
<code>@ai-sdk/*</code>): For streaming AI responses
and multi-provider support
</li>
<li>
<strong>react-drawio</strong>: For diagram
representation and manipulation
</li>
</ul>
<p className="text-gray-700 mt-4">
Diagrams are represented as XML that can be rendered in
draw.io. The AI processes your commands and generates or
modifies this XML accordingly.
</p>
<div className="flex items-start">
<div className="flex-shrink-0 w-10 h-10 bg-blue-600 text-white rounded-full flex items-center justify-center font-bold mr-4">
2
</div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">Describe Your Diagram</h3>
<p className="text-gray-700">
Type your diagram request in natural language. Be as detailed or as general as you like.
You can also upload reference images for the AI to analyze and replicate.
</p>
</div>
</div>
{/* Multi-Provider Support */}
<h2 className="text-2xl font-semibold text-gray-900 mt-10 mb-4">
Multi-Provider Support
</h2>
<ul className="list-disc pl-6 text-gray-700 space-y-1">
<li>AWS Bedrock (default)</li>
<li>
OpenAI / OpenAI-compatible APIs (via{" "}
<code>OPENAI_BASE_URL</code>)
</li>
<li>Anthropic</li>
<li>Google AI</li>
<li>Azure OpenAI</li>
<li>Ollama</li>
<li>OpenRouter</li>
<li>DeepSeek</li>
</ul>
<p className="text-gray-700 mt-4">
Note that <code>claude-sonnet-4-5</code> has trained on
draw.io diagrams with AWS logos, so if you want to
create AWS architecture diagrams, this is the best
choice.
</p>
<div className="flex items-start">
<div className="flex-shrink-0 w-10 h-10 bg-blue-600 text-white rounded-full flex items-center justify-center font-bold mr-4">
3
</div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">AI Generates Your Diagram</h3>
<p className="text-gray-700">
The AI processes your request and automatically creates your diagram in seconds.
Watch as it appears in the editor with proper symbols, layouts, and connections.
</p>
</div>
</div>
{/* Support */}
<div className="flex items-center gap-4 mt-10 mb-4">
<h2 className="text-2xl font-semibold text-gray-900">
Support &amp; Contact
</h2>
<iframe
src="https://github.com/sponsors/DayuanJiang/button"
title="Sponsor DayuanJiang"
height="32"
width="114"
style={{ border: 0, borderRadius: 6 }}
/>
</div>
<p className="text-gray-700">
If you find this project useful, please consider{" "}
<a
href="https://github.com/sponsors/DayuanJiang"
target="_blank"
rel="noopener noreferrer"
className="text-blue-600 hover:underline"
>
sponsoring
</a>{" "}
to help host the live demo site!
</p>
<p className="text-gray-700 mt-2">
For support or inquiries, please open an issue on the{" "}
<a
href="https://github.com/DayuanJiang/next-ai-draw-io"
target="_blank"
rel="noopener noreferrer"
className="text-blue-600 hover:underline"
>
GitHub repository
</a>{" "}
or contact: me[at]jiang.jp
</p>
<div className="flex items-start">
<div className="flex-shrink-0 w-10 h-10 bg-blue-600 text-white rounded-full flex items-center justify-center font-bold mr-4">
4
</div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">Refine and Export</h3>
<p className="text-gray-700">
Request modifications using the chat, manually edit in draw.io, or export to PNG, SVG,
or XML format. Access diagram history to restore previous versions anytime.
</p>
</div>
</div>
</div>
</section>
{/* Benefits */}
<section className="mb-12">
<h2 className="text-2xl font-semibold text-gray-900 mb-6">Why Choose Next AI Draw.io?</h2>
<div className="grid md:grid-cols-2 gap-6">
<div className="flex items-start">
<div className="flex-shrink-0 text-blue-600 text-2xl mr-3"></div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">Save Time</h3>
<p className="text-gray-700">
Create complex diagrams in seconds instead of hours. No more dragging, aligning,
or searching for the right symbolsAI handles it all.
</p>
</div>
</div>
<div className="flex items-start">
<div className="flex-shrink-0 text-blue-600 text-2xl mr-3">🎯</div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">Precision Editing</h3>
<p className="text-gray-700">
Targeted XML editing ensures changes are precise and predictable, unlike tools
that regenerate entire diagrams and lose your layout.
</p>
</div>
</div>
<div className="flex items-start">
<div className="flex-shrink-0 text-blue-600 text-2xl mr-3">🆓</div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">Completely Free</h3>
<p className="text-gray-700">
No subscriptions, no usage limits, no hidden costs. Open-source and free forever.
Use it for personal projects, work, or education.
</p>
</div>
</div>
<div className="flex items-start">
<div className="flex-shrink-0 text-blue-600 text-2xl mr-3">🔒</div>
<div>
<h3 className="text-lg font-semibold text-gray-900 mb-2">Privacy First</h3>
<p className="text-gray-700">
No account required means your diagrams stay private. Work on sensitive
architecture designs without worrying about data storage or privacy policies.
</p>
</div>
</div>
</div>
</section>
{/* CTA Section */}
<section className="bg-blue-600 text-white p-8 rounded-lg text-center">
<h2 className="text-3xl font-bold mb-4">Ready to Create Your First AI Diagram?</h2>
<p className="text-xl mb-6">
Start generating professional diagrams in seconds. No signup required.
</p>
{/* CTA */}
<div className="mt-12 text-center">
<Link
href="/"
className="inline-block bg-white text-blue-600 px-8 py-3 rounded-lg font-semibold hover:bg-gray-100 transition-colors"
className="inline-block bg-blue-600 text-white px-8 py-3 rounded-lg font-semibold hover:bg-blue-700 transition-colors"
>
Open Editor
</Link>
</section>
</div>
</article>
</main>
{/* Footer */}
<footer className="bg-white border-t border-gray-200 mt-16">
<div className="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 py-8">
<div className="text-center text-gray-600 text-sm">
<p className="mb-2">
Next AI Draw.io - Free AI-Powered Diagram Generator
</p>
<p>
Perfect for developers, architects, students, and business analysts.
Open source. No login required.
</p>
</div>
<div className="max-w-4xl mx-auto px-4 sm:px-6 lg:px-8 py-6">
<p className="text-center text-gray-600 text-sm">
Next AI Draw.io - Open Source AI-Powered Diagram
Generator
</p>
</div>
</footer>
</div>
);
)
}

View File

@@ -1,195 +1,568 @@
import { streamText, convertToModelMessages } from 'ai';
import { getAIModel } from '@/lib/ai-providers';
import { z } from "zod";
import {
APICallError,
convertToModelMessages,
createUIMessageStream,
createUIMessageStreamResponse,
InvalidToolInputError,
LoadAPIKeyError,
stepCountIs,
streamText,
} from "ai"
import { jsonrepair } from "jsonrepair"
import { z } from "zod"
import { getAIModel, supportsPromptCaching } from "@/lib/ai-providers"
import { findCachedResponse } from "@/lib/cached-responses"
import {
getTelemetryConfig,
setTraceInput,
setTraceOutput,
wrapWithObserve,
} from "@/lib/langfuse"
import { getSystemPrompt } from "@/lib/system-prompts"
export const maxDuration = 60;
export const maxDuration = 120
export async function POST(req: Request) {
try {
const { messages, xml } = await req.json();
// File upload limits (must match client-side)
const MAX_FILE_SIZE = 2 * 1024 * 1024 // 2MB
const MAX_FILES = 5
const systemMessage = `
You are an expert diagram creation assistant specializing in draw.io XML generation.
Your primary function is crafting clear, well-organized visual diagrams through precise XML specifications.
You can see the image that user uploaded.
Note that when you need to generate diagram about aws architecture, use **AWS 2025 icons**.
// Helper function to validate file parts in messages
function validateFileParts(messages: any[]): {
valid: boolean
error?: string
} {
const lastMessage = messages[messages.length - 1]
const fileParts =
lastMessage?.parts?.filter((p: any) => p.type === "file") || []
You utilize the following tools:
---Tool1---
tool name: display_diagram
description: Display a NEW diagram on draw.io. Use this when creating a diagram from scratch or when major structural changes are needed.
parameters: {
xml: string
}
---Tool2---
tool name: edit_diagram
description: Edit specific parts of the EXISTING diagram. Use this when making small targeted changes like adding/removing elements, changing labels, or adjusting properties. This is more efficient than regenerating the entire diagram.
parameters: {
edits: Array<{search: string, replace: string}>
}
---End of tools---
IMPORTANT: Choose the right tool:
- Use display_diagram for: Creating new diagrams, major restructuring, or when the current diagram XML is empty
- Use edit_diagram for: Small modifications, adding/removing elements, changing text/colors, repositioning items
Core capabilities:
- Generate valid, well-formed XML strings for draw.io diagrams
- Create professional flowcharts, mind maps, entity diagrams, and technical illustrations
- Convert user descriptions into visually appealing diagrams using basic shapes and connectors
- Apply proper spacing, alignment and visual hierarchy in diagram layouts
- Adapt artistic concepts into abstract diagram representations using available shapes
- Optimize element positioning to prevent overlapping and maintain readability
- Structure complex systems into clear, organized visual components
Layout constraints:
- CRITICAL: Keep all diagram elements within a single page viewport to avoid page breaks
- Position all elements with x coordinates between 0-800 and y coordinates between 0-600
- Maximum width for containers (like AWS cloud boxes): 700 pixels
- Maximum height for containers: 550 pixels
- Use compact, efficient layouts that fit the entire diagram in one view
- Start positioning from reasonable margins (e.g., x=40, y=40) and keep elements grouped closely
- For large diagrams with many elements, use vertical stacking or grid layouts that stay within bounds
- Avoid spreading elements too far apart horizontally - users should see the complete diagram without a page break line
Note that:
- Focus on producing clean, professional diagrams that effectively communicate the intended information through thoughtful layout and design choices.
- When artistic drawings are requested, creatively compose them using standard diagram shapes and connectors while maintaining visual clarity.
- Return XML only via tool calls, never in text responses.
- If user asks you to replicate a diagram based on an image, remember to match the diagram style and layout as closely as possible. Especially, pay attention to the lines and shapes, for example, if the lines are straight or curved, and if the shapes are rounded or square.
- Note that when you need to generate diagram about aws architecture, use **AWS 2025 icons**.
When using edit_diagram tool:
- Keep edits minimal - only include the specific line being changed plus 1-2 context lines
- Example GOOD edit: {"search": " <mxCell id=\"2\" value=\"Old Text\">", "replace": " <mxCell id=\"2\" value=\"New Text\">"}
- Example BAD edit: Including 10+ unchanged lines just to change one attribute
- For multiple changes, use separate edits: [{"search": "line1", "replace": "new1"}, {"search": "line2", "replace": "new2"}]
- RETRY POLICY: If edit_diagram fails because the search pattern cannot be found:
* You may retry edit_diagram up to 3 times with adjusted search patterns
* After 3 failed attempts, you MUST fall back to using display_diagram to regenerate the entire diagram
* The error message will indicate how many retries remain
`;
const lastMessage = messages[messages.length - 1];
// Extract text from the last message parts
const lastMessageText = lastMessage.parts?.find((part: any) => part.type === 'text')?.text || '';
// Extract file parts (images) from the last message
const fileParts = lastMessage.parts?.filter((part: any) => part.type === 'file') || [];
const formattedTextContent = `
Current diagram XML:
"""xml
${xml || ''}
"""
User input:
"""md
${lastMessageText}
"""`;
// Convert UIMessages to ModelMessages and add system message
const modelMessages = convertToModelMessages(messages);
let enhancedMessages = [...modelMessages];
// Update the last message with formatted content if it's a user message
if (enhancedMessages.length >= 1) {
const lastModelMessage = enhancedMessages[enhancedMessages.length - 1];
if (lastModelMessage.role === 'user') {
// Build content array with text and file parts
const contentParts: any[] = [
{ type: 'text', text: formattedTextContent }
];
// Add image parts back
for (const filePart of fileParts) {
contentParts.push({
type: 'image',
image: filePart.url,
mimeType: filePart.mediaType
});
if (fileParts.length > MAX_FILES) {
return {
valid: false,
error: `Too many files. Maximum ${MAX_FILES} allowed.`,
}
enhancedMessages = [
...enhancedMessages.slice(0, -1),
{ ...lastModelMessage, content: contentParts }
];
}
}
console.log("Enhanced messages:", enhancedMessages);
for (const filePart of fileParts) {
// Data URLs format: data:image/png;base64,<data>
// Base64 increases size by ~33%, so we check the decoded size
if (filePart.url?.startsWith("data:")) {
const base64Data = filePart.url.split(",")[1]
if (base64Data) {
const sizeInBytes = Math.ceil((base64Data.length * 3) / 4)
if (sizeInBytes > MAX_FILE_SIZE) {
return {
valid: false,
error: `File exceeds ${MAX_FILE_SIZE / 1024 / 1024}MB limit.`,
}
}
}
}
}
// Get AI model from environment configuration
const { model, providerOptions } = getAIModel();
return { valid: true }
}
// Helper function to check if diagram is minimal/empty
function isMinimalDiagram(xml: string): boolean {
const stripped = xml.replace(/\s/g, "")
return !stripped.includes('id="2"')
}
// Helper function to replace historical tool call XML with placeholders
// This reduces token usage and forces LLM to rely on the current diagram XML (source of truth)
function replaceHistoricalToolInputs(messages: any[]): any[] {
return messages.map((msg) => {
if (msg.role !== "assistant" || !Array.isArray(msg.content)) {
return msg
}
const replacedContent = msg.content.map((part: any) => {
if (part.type === "tool-call") {
const toolName = part.toolName
if (
toolName === "display_diagram" ||
toolName === "edit_diagram"
) {
return {
...part,
input: {
placeholder:
"[XML content replaced - see current diagram XML in system context]",
},
}
}
}
return part
})
return { ...msg, content: replacedContent }
})
}
// Helper function to create cached stream response
function createCachedStreamResponse(xml: string): Response {
const toolCallId = `cached-${Date.now()}`
const stream = createUIMessageStream({
execute: async ({ writer }) => {
writer.write({ type: "start" })
writer.write({
type: "tool-input-start",
toolCallId,
toolName: "display_diagram",
})
writer.write({
type: "tool-input-delta",
toolCallId,
inputTextDelta: xml,
})
writer.write({
type: "tool-input-available",
toolCallId,
toolName: "display_diagram",
input: { xml },
})
writer.write({ type: "finish" })
},
})
return createUIMessageStreamResponse({ stream })
}
// Inner handler function
async function handleChatRequest(req: Request): Promise<Response> {
// Check for access code
const accessCodes =
process.env.ACCESS_CODE_LIST?.split(",")
.map((code) => code.trim())
.filter(Boolean) || []
if (accessCodes.length > 0) {
const accessCodeHeader = req.headers.get("x-access-code")
if (!accessCodeHeader || !accessCodes.includes(accessCodeHeader)) {
return Response.json(
{
error: "Invalid or missing access code. Please configure it in Settings.",
},
{ status: 401 },
)
}
}
const { messages, xml, previousXml, sessionId } = await req.json()
// Get user IP for Langfuse tracking
const forwardedFor = req.headers.get("x-forwarded-for")
const userId = forwardedFor?.split(",")[0]?.trim() || "anonymous"
// Validate sessionId for Langfuse (must be string, max 200 chars)
const validSessionId =
sessionId && typeof sessionId === "string" && sessionId.length <= 200
? sessionId
: undefined
// Extract user input text for Langfuse trace
const lastMessage = messages[messages.length - 1]
const userInputText =
lastMessage?.parts?.find((p: any) => p.type === "text")?.text || ""
// Update Langfuse trace with input, session, and user
setTraceInput({
input: userInputText,
sessionId: validSessionId,
userId: userId,
})
// === FILE VALIDATION START ===
const fileValidation = validateFileParts(messages)
if (!fileValidation.valid) {
return Response.json({ error: fileValidation.error }, { status: 400 })
}
// === FILE VALIDATION END ===
// === CACHE CHECK START ===
const isFirstMessage = messages.length === 1
const isEmptyDiagram = !xml || xml.trim() === "" || isMinimalDiagram(xml)
if (isFirstMessage && isEmptyDiagram) {
const lastMessage = messages[0]
const textPart = lastMessage.parts?.find((p: any) => p.type === "text")
const filePart = lastMessage.parts?.find((p: any) => p.type === "file")
const cached = findCachedResponse(textPart?.text || "", !!filePart)
if (cached) {
return createCachedStreamResponse(cached.xml)
}
}
// === CACHE CHECK END ===
// Read client AI provider overrides from headers
const clientOverrides = {
provider: req.headers.get("x-ai-provider"),
baseUrl: req.headers.get("x-ai-base-url"),
apiKey: req.headers.get("x-ai-api-key"),
modelId: req.headers.get("x-ai-model"),
}
// Read minimal style preference from header
const minimalStyle = req.headers.get("x-minimal-style") === "true"
// Get AI model with optional client overrides
const { model, providerOptions, headers, modelId } =
getAIModel(clientOverrides)
// Check if model supports prompt caching
const shouldCache = supportsPromptCaching(modelId)
console.log(
`[Prompt Caching] ${shouldCache ? "ENABLED" : "DISABLED"} for model: ${modelId}`,
)
// Get the appropriate system prompt based on model (extended for Opus/Haiku 4.5)
const systemMessage = getSystemPrompt(modelId, minimalStyle)
// Extract file parts (images) from the last message
const fileParts =
lastMessage.parts?.filter((part: any) => part.type === "file") || []
// User input only - XML is now in a separate cached system message
const formattedUserInput = `User input:
"""md
${userInputText}
"""`
// Convert UIMessages to ModelMessages and add system message
const modelMessages = convertToModelMessages(messages)
// Replace historical tool call XML with placeholders to reduce tokens
// Disabled by default - some models (e.g. minimax) copy placeholders instead of generating XML
const enableHistoryReplace =
process.env.ENABLE_HISTORY_XML_REPLACE === "true"
const placeholderMessages = enableHistoryReplace
? replaceHistoricalToolInputs(modelMessages)
: modelMessages
// Filter out messages with empty content arrays (Bedrock API rejects these)
// This is a safety measure - ideally convertToModelMessages should handle all cases
let enhancedMessages = placeholderMessages.filter(
(msg: any) =>
msg.content && Array.isArray(msg.content) && msg.content.length > 0,
)
// Update the last message with user input only (XML moved to separate cached system message)
if (enhancedMessages.length >= 1) {
const lastModelMessage = enhancedMessages[enhancedMessages.length - 1]
if (lastModelMessage.role === "user") {
// Build content array with user input text and file parts
const contentParts: any[] = [
{ type: "text", text: formattedUserInput },
]
// Add image parts back
for (const filePart of fileParts) {
contentParts.push({
type: "image",
image: filePart.url,
mimeType: filePart.mediaType,
})
}
enhancedMessages = [
...enhancedMessages.slice(0, -1),
{ ...lastModelMessage, content: contentParts },
]
}
}
// Add cache point to the last assistant message in conversation history
// This caches the entire conversation prefix for subsequent requests
// Strategy: system (cached) + history with last assistant (cached) + new user message
if (shouldCache && enhancedMessages.length >= 2) {
// Find the last assistant message (should be second-to-last, before current user message)
for (let i = enhancedMessages.length - 2; i >= 0; i--) {
if (enhancedMessages[i].role === "assistant") {
enhancedMessages[i] = {
...enhancedMessages[i],
providerOptions: {
bedrock: { cachePoint: { type: "default" } },
},
}
break // Only cache the last assistant message
}
}
}
// System messages with multiple cache breakpoints for optimal caching:
// - Breakpoint 1: Static instructions (~1500 tokens) - rarely changes
// - Breakpoint 2: Current XML context - changes per diagram, but constant within a conversation turn
// This allows: if only user message changes, both system caches are reused
// if XML changes, instruction cache is still reused
const systemMessages = [
// Cache breakpoint 1: Instructions (rarely change)
{
role: "system" as const,
content: systemMessage,
...(shouldCache && {
providerOptions: {
bedrock: { cachePoint: { type: "default" } },
},
}),
},
// Cache breakpoint 2: Previous and Current diagram XML context
{
role: "system" as const,
content: `${previousXml ? `Previous diagram XML (before user's last message):\n"""xml\n${previousXml}\n"""\n\n` : ""}Current diagram XML (AUTHORITATIVE - the source of truth):\n"""xml\n${xml || ""}\n"""\n\nIMPORTANT: The "Current diagram XML" is the SINGLE SOURCE OF TRUTH for what's on the canvas right now. The user can manually add, delete, or modify shapes directly in draw.io. Always count and describe elements based on the CURRENT XML, not on what you previously generated. If both previous and current XML are shown, compare them to understand what the user changed. When using edit_diagram, COPY search patterns exactly from the CURRENT XML - attribute order matters!`,
...(shouldCache && {
providerOptions: {
bedrock: { cachePoint: { type: "default" } },
},
}),
},
]
const allMessages = [...systemMessages, ...enhancedMessages]
const result = streamText({
model,
system: systemMessage,
messages: enhancedMessages,
...(providerOptions && { providerOptions }),
tools: {
// Client-side tool that will be executed on the client
display_diagram: {
description: `Display a diagram on draw.io. You only need to pass the nodes inside the <root> tag (including the <root> tag itself) in the XML string.
For example:
<root>
<mxCell id="0"/>
<mxCell id="1" parent="0"/>
<mxGeometry x="20" y="20" width="100" height="100" as="geometry"/>
<mxCell id="2" value="Hello, World!" style="shape=rectangle" parent="1">
<mxGeometry x="20" y="20" width="100" height="100" as="geometry"/>
</mxCell>
</root>
- Note that when you need to generate diagram about aws architecture, use **AWS 2025 icons**.
- If you are asked to generate animated connectors, make sure to include "flowAnimation=1" in the style of the connector elements.
`,
inputSchema: z.object({
xml: z.string().describe("XML string to be displayed on draw.io")
})
model,
...(process.env.MAX_OUTPUT_TOKENS && {
maxOutputTokens: parseInt(process.env.MAX_OUTPUT_TOKENS, 10),
}),
stopWhen: stepCountIs(5),
// Repair truncated tool calls when maxOutputTokens is reached mid-JSON
experimental_repairToolCall: async ({ toolCall, error }) => {
// Only attempt repair for invalid tool input (broken JSON from truncation)
if (
error instanceof InvalidToolInputError ||
error.name === "AI_InvalidToolInputError"
) {
try {
// Use jsonrepair to fix truncated JSON
const repairedInput = jsonrepair(toolCall.input)
console.log(
`[repairToolCall] Repaired truncated JSON for tool: ${toolCall.toolName}`,
)
return { ...toolCall, input: repairedInput }
} catch (repairError) {
console.warn(
`[repairToolCall] Failed to repair JSON for tool: ${toolCall.toolName}`,
repairError,
)
return null
}
}
// Don't attempt to repair other errors (like NoSuchToolError)
return null
},
edit_diagram: {
description: `Edit specific parts of the current diagram by replacing exact line matches. Use this tool to make targeted fixes without regenerating the entire XML.
messages: allMessages,
...(providerOptions && { providerOptions }), // This now includes all reasoning configs
...(headers && { headers }),
// Langfuse telemetry config (returns undefined if not configured)
...(getTelemetryConfig({ sessionId: validSessionId, userId }) && {
experimental_telemetry: getTelemetryConfig({
sessionId: validSessionId,
userId,
}),
}),
onFinish: ({ text, usage }) => {
// Pass usage to Langfuse (Bedrock streaming doesn't auto-report tokens to telemetry)
setTraceOutput(text, {
promptTokens: usage?.inputTokens,
completionTokens: usage?.outputTokens,
})
},
tools: {
// Client-side tool that will be executed on the client
display_diagram: {
description: `Display a diagram on draw.io. Pass ONLY the mxCell elements - wrapper tags and root cells are added automatically.
VALIDATION RULES (XML will be rejected if violated):
1. Generate ONLY mxCell elements - NO wrapper tags (<mxfile>, <mxGraphModel>, <root>)
2. Do NOT include root cells (id="0" or id="1") - they are added automatically
3. All mxCell elements must be siblings - never nested
4. Every mxCell needs a unique id (start from "2")
5. Every mxCell needs a valid parent attribute (use "1" for top-level)
6. Escape special chars in values: &lt; &gt; &amp; &quot;
Example (generate ONLY this - no wrapper tags):
<mxCell id="lane1" value="Frontend" style="swimlane;" vertex="1" parent="1">
<mxGeometry x="40" y="40" width="200" height="200" as="geometry"/>
</mxCell>
<mxCell id="step1" value="Step 1" style="rounded=1;" vertex="1" parent="lane1">
<mxGeometry x="20" y="60" width="160" height="40" as="geometry"/>
</mxCell>
<mxCell id="lane2" value="Backend" style="swimlane;" vertex="1" parent="1">
<mxGeometry x="280" y="40" width="200" height="200" as="geometry"/>
</mxCell>
<mxCell id="step2" value="Step 2" style="rounded=1;" vertex="1" parent="lane2">
<mxGeometry x="20" y="60" width="160" height="40" as="geometry"/>
</mxCell>
<mxCell id="edge1" style="edgeStyle=orthogonalEdgeStyle;endArrow=classic;" edge="1" parent="1" source="step1" target="step2">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
Notes:
- For AWS diagrams, use **AWS 2025 icons**.
- For animated connectors, add "flowAnimation=1" to edge style.
`,
inputSchema: z.object({
xml: z
.string()
.describe("XML string to be displayed on draw.io"),
}),
},
edit_diagram: {
description: `Edit specific parts of the current diagram by replacing exact line matches. Use this tool to make targeted fixes without regenerating the entire XML.
CRITICAL: Copy-paste the EXACT search pattern from the "Current diagram XML" in system context. Do NOT reorder attributes or reformat - the attribute order in draw.io XML varies and you MUST match it exactly.
IMPORTANT: Keep edits concise:
- COPY the exact mxCell line from the current XML (attribute order matters!)
- Only include the lines that are changing, plus 1-2 surrounding lines for context if needed
- Break large changes into multiple smaller edits
- Each search must contain complete lines (never truncate mid-line)
- First match only - be specific enough to target the right element`,
inputSchema: z.object({
edits: z.array(z.object({
search: z.string().describe("Exact lines to search for (including whitespace and indentation)"),
replace: z.string().describe("Replacement lines")
})).describe("Array of search/replace pairs to apply sequentially")
})
- First match only - be specific enough to target the right element
⚠️ JSON ESCAPING: Every " inside string values MUST be escaped as \\". Example: x=\\"100\\" y=\\"200\\" - BOTH quotes need backslashes!`,
inputSchema: z.object({
edits: z
.array(
z.object({
search: z
.string()
.describe(
"EXACT lines copied from current XML (preserve attribute order!)",
),
replace: z
.string()
.describe("Replacement lines"),
}),
)
.describe(
"Array of search/replace pairs to apply sequentially",
),
}),
},
append_diagram: {
description: `Continue generating diagram XML when previous display_diagram output was truncated due to length limits.
WHEN TO USE: Only call this tool after display_diagram was truncated (you'll see an error message about truncation).
CRITICAL INSTRUCTIONS:
1. Do NOT include any wrapper tags - just continue the mxCell elements
2. Continue from EXACTLY where your previous output stopped
3. Complete the remaining mxCell elements
4. If still truncated, call append_diagram again with the next fragment
Example: If previous output ended with '<mxCell id="x" style="rounded=1', continue with ';" vertex="1">...' and complete the remaining elements.`,
inputSchema: z.object({
xml: z
.string()
.describe(
"Continuation XML fragment to append (NO wrapper tags)",
),
}),
},
},
},
temperature: 0,
});
// Error handler function to provide detailed error messages
function errorHandler(error: unknown) {
if (error == null) {
return 'unknown error';
}
if (typeof error === 'string') {
return error;
}
if (error instanceof Error) {
return error.message;
}
return JSON.stringify(error);
}
...(process.env.TEMPERATURE !== undefined && {
temperature: parseFloat(process.env.TEMPERATURE),
}),
})
return result.toUIMessageStreamResponse({
onError: errorHandler,
});
} catch (error) {
console.error('Error in chat route:', error);
return Response.json(
{ error: 'Internal server error' },
{ status: 500 }
);
}
sendReasoning: true,
messageMetadata: ({ part }) => {
if (part.type === "finish") {
const usage = (part as any).totalUsage
if (!usage) {
console.warn(
"[messageMetadata] No usage data in finish part",
)
return undefined
}
// Total input = non-cached + cached (these are separate counts)
// Note: cacheWriteInputTokens is not available on finish part
const totalInputTokens =
(usage.inputTokens ?? 0) + (usage.cachedInputTokens ?? 0)
return {
inputTokens: totalInputTokens,
outputTokens: usage.outputTokens ?? 0,
finishReason: (part as any).finishReason,
}
}
return undefined
},
})
}
// Helper to categorize errors and return appropriate response
function handleError(error: unknown): Response {
console.error("Error in chat route:", error)
const isDev = process.env.NODE_ENV === "development"
// Check for specific AI SDK error types
if (APICallError.isInstance(error)) {
return Response.json(
{
error: error.message,
...(isDev && {
details: error.responseBody,
stack: error.stack,
}),
},
{ status: error.statusCode || 500 },
)
}
if (LoadAPIKeyError.isInstance(error)) {
return Response.json(
{
error: "Authentication failed. Please check your API key.",
...(isDev && {
stack: error.stack,
}),
},
{ status: 401 },
)
}
// Fallback for other errors with safety filter
const message =
error instanceof Error ? error.message : "An unexpected error occurred"
const status = (error as any)?.statusCode || (error as any)?.status || 500
// Prevent leaking API keys, tokens, or other sensitive data
const lowerMessage = message.toLowerCase()
const safeMessage =
lowerMessage.includes("key") ||
lowerMessage.includes("token") ||
lowerMessage.includes("sig") ||
lowerMessage.includes("signature") ||
lowerMessage.includes("secret") ||
lowerMessage.includes("password") ||
lowerMessage.includes("credential")
? "Authentication failed. Please check your credentials."
: message
return Response.json(
{
error: safeMessage,
...(isDev && {
details: message,
stack: error instanceof Error ? error.stack : undefined,
}),
},
{ status },
)
}
// Wrap handler with error handling
async function safeHandler(req: Request): Promise<Response> {
try {
return await handleChatRequest(req)
} catch (error) {
return handleError(error)
}
}
// Wrap with Langfuse observe (if configured)
const observedHandler = wrapWithObserve(safeHandler)
export async function POST(req: Request) {
return observedHandler(req)
}

View File

@@ -81,16 +81,15 @@ Contains the actual diagram data.
## Root Cell Container: `<root>`
Contains all the cells in the diagram.
Contains all the cells in the diagram. **Note:** When generating diagrams, you only need to provide the mxCell elements - the root container and root cells (id="0", id="1") are added automatically.
**Example:**
**Internal structure (auto-generated):**
```xml
<root>
<mxCell id="0"/>
<mxCell id="1" parent="0"/>
<!-- Other cells go here -->
<mxCell id="0"/> <!-- Auto-added -->
<mxCell id="1" parent="0"/> <!-- Auto-added -->
<!-- Your mxCell elements go here (start from id="2") -->
</root>
```
@@ -203,14 +202,15 @@ Draw.io files contain two special cells that are always present:
1. **Root Cell** (id = "0"): The parent of all cells
2. **Default Parent Cell** (id = "1", parent = "0"): The default layer and parent for most cells
## Tips for Manually Creating Draw.io XML
## Tips for Creating Draw.io XML
1. Start with the basic structure (`mxfile`, `diagram`, `mxGraphModel`, `root`)
2. Always include the two special cells (id = "0" and id = "1")
1. **Generate ONLY mxCell elements** - wrapper tags and root cells (id="0", id="1") are added automatically
2. Start IDs from "2" (id="0" and id="1" are reserved for root cells)
3. Assign unique and sequential IDs to all cells
4. Define parent relationships correctly
4. Define parent relationships correctly (use parent="1" for top-level shapes)
5. Use `mxGeometry` elements to position shapes
6. For connectors, specify `source` and `target` attributes
7. **CRITICAL: All mxCell elements must be siblings. NEVER nest mxCell inside another mxCell.**
## Common Patterns
@@ -234,12 +234,33 @@ To group elements, create a parent cell and set other cells' `parent` attribute
### Swimlanes
Swimlanes use the `swimlane` shape style:
Swimlanes use the `swimlane` shape style. **IMPORTANT: All mxCell elements (swimlanes, steps, and edges) must be siblings under `<root>`. Edges are NOT nested inside swimlanes or steps.**
```xml
<mxCell id="20" value="Swimlane 1" style="swimlane;fontStyle=0;childLayout=stackLayout;horizontal=1;startSize=30;horizontalStack=0;resizeParent=1;resizeParentMax=0;resizeLast=0;collapsible=1;marginBottom=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
<mxGeometry x="200" y="200" width="140" height="120" as="geometry" />
</mxCell>
<root>
<mxCell id="0"/>
<mxCell id="1" parent="0"/>
<!-- Swimlane 1 -->
<mxCell id="lane1" value="Frontend" style="swimlane;startSize=30;" vertex="1" parent="1">
<mxGeometry x="40" y="40" width="200" height="300" as="geometry"/>
</mxCell>
<!-- Swimlane 2 -->
<mxCell id="lane2" value="Backend" style="swimlane;startSize=30;" vertex="1" parent="1">
<mxGeometry x="280" y="40" width="200" height="300" as="geometry"/>
</mxCell>
<!-- Step inside lane1 (parent="lane1") -->
<mxCell id="step1" value="Send Request" style="rounded=1;" vertex="1" parent="lane1">
<mxGeometry x="20" y="60" width="160" height="40" as="geometry"/>
</mxCell>
<!-- Step inside lane2 (parent="lane2") -->
<mxCell id="step2" value="Process" style="rounded=1;" vertex="1" parent="lane2">
<mxGeometry x="20" y="60" width="160" height="40" as="geometry"/>
</mxCell>
<!-- Edge connecting step1 to step2 (sibling element, NOT nested inside steps) -->
<mxCell id="edge1" style="edgeStyle=orthogonalEdgeStyle;endArrow=classic;" edge="1" parent="1" source="step1" target="step2">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
</root>
```
### Tables

10
app/api/config/route.ts Normal file
View File

@@ -0,0 +1,10 @@
import { NextResponse } from "next/server"
export async function GET() {
return NextResponse.json({
accessCodeRequired: !!process.env.ACCESS_CODE_LIST,
dailyRequestLimit: Number(process.env.DAILY_REQUEST_LIMIT) || 0,
dailyTokenLimit: Number(process.env.DAILY_TOKEN_LIMIT) || 0,
tpmLimit: Number(process.env.TPM_LIMIT) || 0,
})
}

View File

@@ -0,0 +1,112 @@
import { randomUUID } from "crypto"
import { z } from "zod"
import { getLangfuseClient } from "@/lib/langfuse"
const feedbackSchema = z.object({
messageId: z.string().min(1).max(200),
feedback: z.enum(["good", "bad"]),
sessionId: z.string().min(1).max(200).optional(),
})
export async function POST(req: Request) {
const langfuse = getLangfuseClient()
if (!langfuse) {
return Response.json({ success: true, logged: false })
}
// Validate input
let data
try {
data = feedbackSchema.parse(await req.json())
} catch {
return Response.json(
{ success: false, error: "Invalid input" },
{ status: 400 },
)
}
const { messageId, feedback, sessionId } = data
// Get user IP for tracking
const forwardedFor = req.headers.get("x-forwarded-for")
const userId = forwardedFor?.split(",")[0]?.trim() || "anonymous"
try {
// Find the most recent chat trace for this session to attach the score to
const tracesResponse = await langfuse.api.trace.list({
sessionId,
limit: 1,
})
const traces = tracesResponse.data || []
const latestTrace = traces[0]
if (!latestTrace) {
// No trace found for this session - create a standalone feedback trace
const traceId = randomUUID()
const timestamp = new Date().toISOString()
await langfuse.api.ingestion.batch({
batch: [
{
type: "trace-create",
id: randomUUID(),
timestamp,
body: {
id: traceId,
name: "user-feedback",
sessionId,
userId,
input: { messageId, feedback },
metadata: {
source: "feedback-button",
note: "standalone - no chat trace found",
},
timestamp,
},
},
{
type: "score-create",
id: randomUUID(),
timestamp,
body: {
id: randomUUID(),
traceId,
name: "user-feedback",
value: feedback === "good" ? 1 : 0,
comment: `User gave ${feedback} feedback`,
},
},
],
})
} else {
// Attach score to the existing chat trace
const timestamp = new Date().toISOString()
await langfuse.api.ingestion.batch({
batch: [
{
type: "score-create",
id: randomUUID(),
timestamp,
body: {
id: randomUUID(),
traceId: latestTrace.id,
name: "user-feedback",
value: feedback === "good" ? 1 : 0,
comment: `User gave ${feedback} feedback`,
},
},
],
})
}
return Response.json({ success: true, logged: true })
} catch (error) {
console.error("Langfuse feedback error:", error)
return Response.json(
{ success: false, error: "Failed to log feedback" },
{ status: 500 },
)
}
}

71
app/api/log-save/route.ts Normal file
View File

@@ -0,0 +1,71 @@
import { randomUUID } from "crypto"
import { z } from "zod"
import { getLangfuseClient } from "@/lib/langfuse"
const saveSchema = z.object({
filename: z.string().min(1).max(255),
format: z.enum(["drawio", "png", "svg"]),
sessionId: z.string().min(1).max(200).optional(),
})
export async function POST(req: Request) {
const langfuse = getLangfuseClient()
if (!langfuse) {
return Response.json({ success: true, logged: false })
}
// Validate input
let data
try {
data = saveSchema.parse(await req.json())
} catch {
return Response.json(
{ success: false, error: "Invalid input" },
{ status: 400 },
)
}
const { filename, format, sessionId } = data
try {
const timestamp = new Date().toISOString()
// Find the most recent chat trace for this session to attach the save flag
const tracesResponse = await langfuse.api.trace.list({
sessionId,
limit: 1,
})
const traces = tracesResponse.data || []
const latestTrace = traces[0]
if (latestTrace) {
// Add a score to the existing trace to flag that user saved
await langfuse.api.ingestion.batch({
batch: [
{
type: "score-create",
id: randomUUID(),
timestamp,
body: {
id: randomUUID(),
traceId: latestTrace.id,
name: "diagram-saved",
value: 1,
comment: `User saved diagram as ${filename}.${format}`,
},
},
],
})
}
// If no trace found, skip logging (user hasn't chatted yet)
return Response.json({ success: true, logged: !!latestTrace })
} catch (error) {
console.error("Langfuse save error:", error)
return Response.json(
{ success: false, error: "Failed to log save" },
{ status: 500 },
)
}
}

View File

@@ -0,0 +1,32 @@
export async function POST(req: Request) {
const accessCodes =
process.env.ACCESS_CODE_LIST?.split(",")
.map((code) => code.trim())
.filter(Boolean) || []
// If no access codes configured, verification always passes
if (accessCodes.length === 0) {
return Response.json({
valid: true,
message: "No access code required",
})
}
const accessCodeHeader = req.headers.get("x-access-code")
if (!accessCodeHeader) {
return Response.json(
{ valid: false, message: "Access code is required" },
{ status: 401 },
)
}
if (!accessCodes.includes(accessCodeHeader)) {
return Response.json(
{ valid: false, message: "Invalid access code" },
{ status: 401 },
)
}
return Response.json({ valid: true, message: "Access code is valid" })
}

View File

@@ -1,123 +1,259 @@
@import "tailwindcss";
@plugin "tailwindcss-animate";
@plugin "@tailwindcss/typography";
@custom-variant dark (&:is(.dark *));
@theme inline {
--color-background: var(--background);
--color-foreground: var(--foreground);
--font-sans: var(--font-geist-sans);
--font-mono: var(--font-geist-mono);
--color-sidebar-ring: var(--sidebar-ring);
--color-sidebar-border: var(--sidebar-border);
--color-sidebar-accent-foreground: var(--sidebar-accent-foreground);
--color-sidebar-accent: var(--sidebar-accent);
--color-sidebar-primary-foreground: var(--sidebar-primary-foreground);
--color-sidebar-primary: var(--sidebar-primary);
--color-sidebar-foreground: var(--sidebar-foreground);
--color-sidebar: var(--sidebar);
--color-chart-5: var(--chart-5);
--color-chart-4: var(--chart-4);
--color-chart-3: var(--chart-3);
--color-chart-2: var(--chart-2);
--color-chart-1: var(--chart-1);
--color-ring: var(--ring);
--color-input: var(--input);
--color-border: var(--border);
--color-destructive: var(--destructive);
--color-accent-foreground: var(--accent-foreground);
--color-accent: var(--accent);
--color-muted-foreground: var(--muted-foreground);
--color-muted: var(--muted);
--color-secondary-foreground: var(--secondary-foreground);
--color-secondary: var(--secondary);
--color-primary-foreground: var(--primary-foreground);
--color-primary: var(--primary);
--color-popover-foreground: var(--popover-foreground);
--color-popover: var(--popover);
--color-card-foreground: var(--card-foreground);
--color-card: var(--card);
--radius-sm: calc(var(--radius) - 4px);
--radius-md: calc(var(--radius) - 2px);
--radius-lg: var(--radius);
--radius-xl: calc(var(--radius) + 4px);
--color-background: var(--background);
--color-foreground: var(--foreground);
--font-sans: var(--font-sans);
--font-mono: var(--font-mono);
--color-sidebar-ring: var(--sidebar-ring);
--color-sidebar-border: var(--sidebar-border);
--color-sidebar-accent-foreground: var(--sidebar-accent-foreground);
--color-sidebar-accent: var(--sidebar-accent);
--color-sidebar-primary-foreground: var(--sidebar-primary-foreground);
--color-sidebar-primary: var(--sidebar-primary);
--color-sidebar-foreground: var(--sidebar-foreground);
--color-sidebar: var(--sidebar);
--color-chart-5: var(--chart-5);
--color-chart-4: var(--chart-4);
--color-chart-3: var(--chart-3);
--color-chart-2: var(--chart-2);
--color-chart-1: var(--chart-1);
--color-ring: var(--ring);
--color-input: var(--input);
--color-border: var(--border);
--color-destructive: var(--destructive);
--color-accent-foreground: var(--accent-foreground);
--color-accent: var(--accent);
--color-muted-foreground: var(--muted-foreground);
--color-muted: var(--muted);
--color-secondary-foreground: var(--secondary-foreground);
--color-secondary: var(--secondary);
--color-primary-foreground: var(--primary-foreground);
--color-primary: var(--primary);
--color-popover-foreground: var(--popover-foreground);
--color-popover: var(--popover);
--color-card-foreground: var(--card-foreground);
--color-card: var(--card);
--radius-sm: calc(var(--radius) - 4px);
--radius-md: calc(var(--radius) - 2px);
--radius-lg: var(--radius);
--radius-xl: calc(var(--radius) + 4px);
}
:root {
--radius: 0.625rem;
--background: oklch(1 0 0);
--foreground: oklch(0.145 0 0);
--card: oklch(1 0 0);
--card-foreground: oklch(0.145 0 0);
--popover: oklch(1 0 0);
--popover-foreground: oklch(0.145 0 0);
--primary: oklch(0.205 0 0);
--primary-foreground: oklch(0.985 0 0);
--secondary: oklch(0.97 0 0);
--secondary-foreground: oklch(0.205 0 0);
--muted: oklch(0.97 0 0);
--muted-foreground: oklch(0.556 0 0);
--accent: oklch(0.97 0 0);
--accent-foreground: oklch(0.205 0 0);
--destructive: oklch(0.577 0.245 27.325);
--border: oklch(0.922 0 0);
--input: oklch(0.922 0 0);
--ring: oklch(0.708 0 0);
--chart-1: oklch(0.646 0.222 41.116);
--chart-2: oklch(0.6 0.118 184.704);
--chart-3: oklch(0.398 0.07 227.392);
--chart-4: oklch(0.828 0.189 84.429);
--chart-5: oklch(0.769 0.188 70.08);
--sidebar: oklch(0.985 0 0);
--sidebar-foreground: oklch(0.145 0 0);
--sidebar-primary: oklch(0.205 0 0);
--sidebar-primary-foreground: oklch(0.985 0 0);
--sidebar-accent: oklch(0.97 0 0);
--sidebar-accent-foreground: oklch(0.205 0 0);
--sidebar-border: oklch(0.922 0 0);
--sidebar-ring: oklch(0.708 0 0);
--radius: 0.75rem;
/* Clean Light Modern Palette */
--background: oklch(0.985 0.002 240);
--foreground: oklch(0.23 0.02 260);
--card: oklch(1 0 0);
--card-foreground: oklch(0.23 0.02 260);
--popover: oklch(1 0 0);
--popover-foreground: oklch(0.23 0.02 260);
/* Dark primary - slightly lighter */
--primary: oklch(0.35 0.01 260);
--primary-foreground: oklch(0.99 0 0);
/* Warm gray secondary */
--secondary: oklch(0.96 0.005 260);
--secondary-foreground: oklch(0.35 0.02 260);
/* Light muted tones */
--muted: oklch(0.965 0.005 260);
--muted-foreground: oklch(0.5 0.02 260);
/* Soft lavender accent */
--accent: oklch(0.94 0.03 280);
--accent-foreground: oklch(0.35 0.08 270);
/* Coral destructive */
--destructive: oklch(0.6 0.2 25);
/* Subtle borders */
--border: oklch(0.92 0.01 260);
--input: oklch(0.94 0.01 260);
--ring: oklch(0.25 0.01 260);
/* Chart colors - harmonious palette */
--chart-1: oklch(0.55 0.18 265);
--chart-2: oklch(0.65 0.15 170);
--chart-3: oklch(0.7 0.18 45);
--chart-4: oklch(0.6 0.2 330);
--chart-5: oklch(0.5 0.15 200);
/* Sidebar */
--sidebar: oklch(0.99 0.002 260);
--sidebar-foreground: oklch(0.23 0.02 260);
--sidebar-primary: oklch(0.55 0.18 265);
--sidebar-primary-foreground: oklch(0.99 0 0);
--sidebar-accent: oklch(0.96 0.02 270);
--sidebar-accent-foreground: oklch(0.35 0.05 265);
--sidebar-border: oklch(0.93 0.01 260);
--sidebar-ring: oklch(0.55 0.18 265);
}
.dark {
--background: oklch(0.145 0 0);
--foreground: oklch(0.985 0 0);
--card: oklch(0.205 0 0);
--card-foreground: oklch(0.985 0 0);
--popover: oklch(0.205 0 0);
--popover-foreground: oklch(0.985 0 0);
--primary: oklch(0.922 0 0);
--primary-foreground: oklch(0.205 0 0);
--secondary: oklch(0.269 0 0);
--secondary-foreground: oklch(0.985 0 0);
--muted: oklch(0.269 0 0);
--muted-foreground: oklch(0.708 0 0);
--accent: oklch(0.269 0 0);
--accent-foreground: oklch(0.985 0 0);
--destructive: oklch(0.704 0.191 22.216);
--border: oklch(1 0 0 / 10%);
--input: oklch(1 0 0 / 15%);
--ring: oklch(0.556 0 0);
--chart-1: oklch(0.488 0.243 264.376);
--chart-2: oklch(0.696 0.17 162.48);
--chart-3: oklch(0.769 0.188 70.08);
--chart-4: oklch(0.627 0.265 303.9);
--chart-5: oklch(0.645 0.246 16.439);
--sidebar: oklch(0.205 0 0);
--sidebar-foreground: oklch(0.985 0 0);
--sidebar-primary: oklch(0.488 0.243 264.376);
--sidebar-primary-foreground: oklch(0.985 0 0);
--sidebar-accent: oklch(0.269 0 0);
--sidebar-accent-foreground: oklch(0.985 0 0);
--sidebar-border: oklch(1 0 0 / 10%);
--sidebar-ring: oklch(0.556 0 0);
--background: oklch(0.15 0.015 260);
--foreground: oklch(0.95 0.01 260);
--card: oklch(0.2 0.015 260);
--card-foreground: oklch(0.95 0.01 260);
--popover: oklch(0.2 0.015 260);
--popover-foreground: oklch(0.95 0.01 260);
--primary: oklch(0.7 0.16 265);
--primary-foreground: oklch(0.15 0.02 260);
--secondary: oklch(0.25 0.015 260);
--secondary-foreground: oklch(0.9 0.01 260);
--muted: oklch(0.25 0.015 260);
--muted-foreground: oklch(0.65 0.02 260);
--accent: oklch(0.3 0.04 280);
--accent-foreground: oklch(0.9 0.03 270);
--destructive: oklch(0.65 0.22 25);
--border: oklch(0.28 0.015 260);
--input: oklch(0.25 0.015 260);
--ring: oklch(0.7 0.16 265);
--chart-1: oklch(0.7 0.16 265);
--chart-2: oklch(0.7 0.13 170);
--chart-3: oklch(0.75 0.16 45);
--chart-4: oklch(0.7 0.18 330);
--chart-5: oklch(0.6 0.13 200);
--sidebar: oklch(0.18 0.015 260);
--sidebar-foreground: oklch(0.95 0.01 260);
--sidebar-primary: oklch(0.7 0.16 265);
--sidebar-primary-foreground: oklch(0.15 0.02 260);
--sidebar-accent: oklch(0.25 0.03 270);
--sidebar-accent-foreground: oklch(0.9 0.02 265);
--sidebar-border: oklch(0.28 0.015 260);
--sidebar-ring: oklch(0.7 0.16 265);
}
@layer base {
* {
@apply border-border outline-ring/50;
}
body {
@apply bg-background text-foreground;
}
* {
@apply border-border outline-ring/50;
}
body {
@apply bg-background text-foreground font-sans;
}
}
/* Fix for Radix ScrollArea viewport horizontal overflow */
[data-slot="scroll-area-viewport"] > div {
display: block !important;
width: 100% !important;
}
/* Custom scrollbar */
@layer utilities {
.scrollbar-thin {
scrollbar-width: thin;
scrollbar-color: oklch(0.85 0.01 260) transparent;
}
.scrollbar-thin::-webkit-scrollbar {
width: 6px;
}
.scrollbar-thin::-webkit-scrollbar-track {
background: transparent;
}
.scrollbar-thin::-webkit-scrollbar-thumb {
background-color: oklch(0.85 0.01 260);
border-radius: 3px;
}
.scrollbar-thin::-webkit-scrollbar-thumb:hover {
background-color: oklch(0.75 0.01 260);
}
}
/* Smooth page transitions */
@keyframes fadeIn {
from {
opacity: 0;
transform: translateY(8px);
}
to {
opacity: 1;
transform: translateY(0);
}
}
@keyframes slideInRight {
from {
opacity: 0;
transform: translateX(16px);
}
to {
opacity: 1;
transform: translateX(0);
}
}
.animate-fade-in {
animation: fadeIn 0.3s ease-out forwards;
}
.animate-slide-in-right {
animation: slideInRight 0.3s ease-out forwards;
}
/* Message bubble animations */
@keyframes messageIn {
from {
opacity: 0;
transform: translateY(12px) scale(0.98);
}
to {
opacity: 1;
transform: translateY(0) scale(1);
}
}
.animate-message-in {
animation: messageIn 0.25s ease-out forwards;
}
/* Subtle floating shadow for cards */
.shadow-soft {
box-shadow:
0 1px 2px oklch(0.23 0.02 260 / 0.04),
0 4px 12px oklch(0.23 0.02 260 / 0.06),
0 8px 24px oklch(0.23 0.02 260 / 0.04);
}
.shadow-soft-lg {
box-shadow:
0 2px 4px oklch(0.23 0.02 260 / 0.04),
0 8px 20px oklch(0.23 0.02 260 / 0.08),
0 16px 40px oklch(0.23 0.02 260 / 0.06);
}
/* Gradient text utility */
.text-gradient-primary {
background: linear-gradient(
135deg,
oklch(0.55 0.18 265),
oklch(0.6 0.2 290)
);
-webkit-background-clip: text;
-webkit-text-fill-color: transparent;
background-clip: text;
}

View File

@@ -1,31 +1,52 @@
import type { Metadata } from "next";
import { Geist, Geist_Mono } from "next/font/google";
import { Analytics } from "@vercel/analytics/react";
import { DiagramProvider } from "@/contexts/diagram-context";
import { GoogleAnalytics } from "@next/third-parties/google"
import type { Metadata, Viewport } from "next"
import { JetBrains_Mono, Plus_Jakarta_Sans } from "next/font/google"
import { DiagramProvider } from "@/contexts/diagram-context"
import "./globals.css";
import "./globals.css"
const geistSans = Geist({
variable: "--font-geist-sans",
const plusJakarta = Plus_Jakarta_Sans({
variable: "--font-sans",
subsets: ["latin"],
});
weight: ["400", "500", "600", "700"],
})
const geistMono = Geist_Mono({
variable: "--font-geist-mono",
const jetbrainsMono = JetBrains_Mono({
variable: "--font-mono",
subsets: ["latin"],
});
weight: ["400", "500"],
})
export const viewport: Viewport = {
width: "device-width",
initialScale: 1,
maximumScale: 1,
userScalable: false,
}
export const metadata: Metadata = {
title: "Next AI Draw.io - AI-Powered Diagram Generator",
description: "Create AWS architecture diagrams, flowcharts, and technical diagrams using AI. Free online tool integrating draw.io with AI assistance for professional diagram creation.",
keywords: ["AI diagram generator", "AWS architecture", "flowchart creator", "draw.io", "AI drawing tool", "technical diagrams", "diagram automation", "free diagram generator", "online diagram maker"],
description:
"Create AWS architecture diagrams, flowcharts, and technical diagrams using AI. Free online tool integrating draw.io with AI assistance for professional diagram creation.",
keywords: [
"AI diagram generator",
"AWS architecture",
"flowchart creator",
"draw.io",
"AI drawing tool",
"technical diagrams",
"diagram automation",
"free diagram generator",
"online diagram maker",
],
authors: [{ name: "Next AI Draw.io" }],
creator: "Next AI Draw.io",
publisher: "Next AI Draw.io",
metadataBase: new URL("https://next-ai-drawio.jiang.jp"),
openGraph: {
title: "Next AI Draw.io - AI Diagram Generator",
description: "Create professional diagrams with AI assistance. Supports AWS architecture, flowcharts, and more.",
description:
"Create professional diagrams with AI assistance. Supports AWS architecture, flowcharts, and more.",
type: "website",
url: "https://next-ai-drawio.jiang.jp",
siteName: "Next AI Draw.io",
@@ -42,7 +63,8 @@ export const metadata: Metadata = {
twitter: {
card: "summary_large_image",
title: "Next AI Draw.io - AI Diagram Generator",
description: "Create professional diagrams with AI assistance. Free, no login required.",
description:
"Create professional diagrams with AI assistance. Free, no login required.",
images: ["/architecture.png"],
},
robots: {
@@ -59,30 +81,31 @@ export const metadata: Metadata = {
icons: {
icon: "/favicon.ico",
},
};
}
export default function RootLayout({
children,
}: Readonly<{
children: React.ReactNode;
children: React.ReactNode
}>) {
const jsonLd = {
'@context': 'https://schema.org',
'@type': 'SoftwareApplication',
name: 'Next AI Draw.io',
applicationCategory: 'DesignApplication',
operatingSystem: 'Web Browser',
description: 'AI-powered diagram generator with targeted XML editing capabilities that integrates with draw.io for creating AWS architecture diagrams, flowcharts, and technical diagrams. Features diagram history, multi-provider AI support, and real-time collaboration.',
url: 'https://next-ai-drawio.jiang.jp',
"@context": "https://schema.org",
"@type": "SoftwareApplication",
name: "Next AI Draw.io",
applicationCategory: "DesignApplication",
operatingSystem: "Web Browser",
description:
"AI-powered diagram generator with targeted XML editing capabilities that integrates with draw.io for creating AWS architecture diagrams, flowcharts, and technical diagrams. Features diagram history, multi-provider AI support, and real-time collaboration.",
url: "https://next-ai-drawio.jiang.jp",
offers: {
'@type': 'Offer',
price: '0',
priceCurrency: 'USD',
"@type": "Offer",
price: "0",
priceCurrency: "USD",
},
};
}
return (
<html lang="en">
<html lang="en" suppressHydrationWarning>
<head>
<script
type="application/ld+json"
@@ -90,12 +113,13 @@ export default function RootLayout({
/>
</head>
<body
className={`${geistSans.variable} ${geistMono.variable} antialiased`}
className={`${plusJakarta.variable} ${jetbrainsMono.variable} antialiased`}
>
<DiagramProvider>{children}</DiagramProvider>
<Analytics />
</body>
{process.env.NEXT_PUBLIC_GA_ID && (
<GoogleAnalytics gaId={process.env.NEXT_PUBLIC_GA_ID} />
)}
</html>
);
)
}

View File

@@ -1,76 +1,202 @@
"use client";
import React, { useState, useEffect } from "react";
import { DrawIoEmbed } from "react-drawio";
import ChatPanel from "@/components/chat-panel";
import { useDiagram } from "@/contexts/diagram-context";
"use client"
import { useEffect, useRef, useState } from "react"
import { DrawIoEmbed } from "react-drawio"
import type { ImperativePanelHandle } from "react-resizable-panels"
import ChatPanel from "@/components/chat-panel"
import { STORAGE_CLOSE_PROTECTION_KEY } from "@/components/settings-dialog"
import {
ResizableHandle,
ResizablePanel,
ResizablePanelGroup,
} from "@/components/ui/resizable"
import { useDiagram } from "@/contexts/diagram-context"
const drawioBaseUrl =
process.env.NEXT_PUBLIC_DRAWIO_BASE_URL || "https://embed.diagrams.net"
export default function Home() {
const { drawioRef, handleDiagramExport } = useDiagram();
const [isMobile, setIsMobile] = useState(false);
const [isChatVisible, setIsChatVisible] = useState(true);
const { drawioRef, handleDiagramExport, onDrawioLoad, resetDrawioReady } =
useDiagram()
const [isMobile, setIsMobile] = useState(false)
const [isChatVisible, setIsChatVisible] = useState(true)
const [drawioUi, setDrawioUi] = useState<"min" | "sketch">("min")
const [darkMode, setDarkMode] = useState(false)
const [isLoaded, setIsLoaded] = useState(false)
const [closeProtection, setCloseProtection] = useState(false)
const chatPanelRef = useRef<ImperativePanelHandle>(null)
// Load preferences from localStorage after mount
useEffect(() => {
const savedUi = localStorage.getItem("drawio-theme")
if (savedUi === "min" || savedUi === "sketch") {
setDrawioUi(savedUi)
}
const savedDarkMode = localStorage.getItem("next-ai-draw-io-dark-mode")
if (savedDarkMode !== null) {
// Use saved preference
const isDark = savedDarkMode === "true"
setDarkMode(isDark)
document.documentElement.classList.toggle("dark", isDark)
} else {
// First visit: match browser preference
const prefersDark = window.matchMedia(
"(prefers-color-scheme: dark)",
).matches
setDarkMode(prefersDark)
document.documentElement.classList.toggle("dark", prefersDark)
}
const savedCloseProtection = localStorage.getItem(
STORAGE_CLOSE_PROTECTION_KEY,
)
if (savedCloseProtection === "true") {
setCloseProtection(true)
}
setIsLoaded(true)
}, [])
const toggleDarkMode = () => {
const newValue = !darkMode
setDarkMode(newValue)
localStorage.setItem("next-ai-draw-io-dark-mode", String(newValue))
document.documentElement.classList.toggle("dark", newValue)
// Reset so onDrawioLoad fires again after remount
resetDrawioReady()
}
// Check mobile
useEffect(() => {
const checkMobile = () => {
setIsMobile(window.innerWidth < 768);
};
setIsMobile(window.innerWidth < 768)
}
// Check on mount
checkMobile();
checkMobile()
window.addEventListener("resize", checkMobile)
return () => window.removeEventListener("resize", checkMobile)
}, [])
// Add event listener for resize
window.addEventListener("resize", checkMobile);
const toggleChatPanel = () => {
const panel = chatPanelRef.current
if (panel) {
if (panel.isCollapsed()) {
panel.expand()
setIsChatVisible(true)
} else {
panel.collapse()
setIsChatVisible(false)
}
}
}
// Cleanup
return () => window.removeEventListener("resize", checkMobile);
}, []);
// Add keyboard shortcut for toggling chat panel (Ctrl+B)
// Keyboard shortcut for toggling chat panel
useEffect(() => {
const handleKeyDown = (event: KeyboardEvent) => {
if ((event.ctrlKey || event.metaKey) && event.key === 'b') {
event.preventDefault();
setIsChatVisible((prev) => !prev);
if ((event.ctrlKey || event.metaKey) && event.key === "b") {
event.preventDefault()
toggleChatPanel()
}
};
}
window.addEventListener('keydown', handleKeyDown);
window.addEventListener("keydown", handleKeyDown)
return () => window.removeEventListener("keydown", handleKeyDown)
}, [])
return () => {
window.removeEventListener('keydown', handleKeyDown);
};
}, []);
// Show confirmation dialog when user tries to leave the page
useEffect(() => {
if (!closeProtection) return
const handleBeforeUnload = (event: BeforeUnloadEvent) => {
event.preventDefault()
return ""
}
window.addEventListener("beforeunload", handleBeforeUnload)
return () =>
window.removeEventListener("beforeunload", handleBeforeUnload)
}, [closeProtection])
return (
<div className="flex h-screen bg-gray-100 relative">
{/* Mobile warning overlay - keeps components mounted */}
{isMobile && (
<div className="absolute inset-0 z-50 flex items-center justify-center bg-gray-100">
<div className="text-center p-8">
<h1 className="text-2xl font-semibold text-gray-800">
Please open this application on a desktop or laptop
</h1>
<div className="h-screen bg-background relative overflow-hidden">
<ResizablePanelGroup
id="main-panel-group"
key={isMobile ? "mobile" : "desktop"}
direction={isMobile ? "vertical" : "horizontal"}
className="h-full"
>
{/* Draw.io Canvas */}
<ResizablePanel
id="drawio-panel"
defaultSize={isMobile ? 50 : 67}
minSize={20}
>
<div
className={`h-full relative ${
isMobile ? "p-1" : "p-2"
}`}
>
<div className="h-full rounded-xl overflow-hidden shadow-soft-lg border border-border/30">
{isLoaded ? (
<DrawIoEmbed
key={`${drawioUi}-${darkMode}`}
ref={drawioRef}
onExport={handleDiagramExport}
onLoad={onDrawioLoad}
baseUrl={drawioBaseUrl}
urlParameters={{
ui: drawioUi,
spin: true,
libraries: false,
saveAndExit: false,
noExitBtn: true,
dark: darkMode,
}}
/>
) : (
<div className="h-full w-full flex items-center justify-center bg-background">
<div className="animate-spin h-8 w-8 border-4 border-primary border-t-transparent rounded-full" />
</div>
)}
</div>
</div>
</div>
)}
</ResizablePanel>
<div className={`${isChatVisible ? 'w-2/3' : 'w-full'} p-1 h-full relative transition-all duration-300 ease-in-out`}>
<DrawIoEmbed
ref={drawioRef}
onExport={handleDiagramExport}
urlParameters={{
spin: true,
libraries: false,
saveAndExit: false,
noExitBtn: true,
}}
/>
</div>
<div className={`${isChatVisible ? 'w-1/3' : 'w-12'} h-full p-1 transition-all duration-300 ease-in-out`}>
<ChatPanel
isVisible={isChatVisible}
onToggleVisibility={() => setIsChatVisible(!isChatVisible)}
/>
</div>
<ResizableHandle withHandle />
{/* Chat Panel */}
<ResizablePanel
id="chat-panel"
ref={chatPanelRef}
defaultSize={isMobile ? 50 : 33}
minSize={isMobile ? 20 : 15}
maxSize={isMobile ? 80 : 50}
collapsible={!isMobile}
collapsedSize={isMobile ? 0 : 3}
onCollapse={() => setIsChatVisible(false)}
onExpand={() => setIsChatVisible(true)}
>
<div className={`h-full ${isMobile ? "p-1" : "py-2 pr-2"}`}>
<ChatPanel
isVisible={isChatVisible}
onToggleVisibility={toggleChatPanel}
drawioUi={drawioUi}
onToggleDrawioUi={() => {
const newUi =
drawioUi === "min" ? "sketch" : "min"
localStorage.setItem("drawio-theme", newUi)
setDrawioUi(newUi)
resetDrawioReady()
}}
darkMode={darkMode}
onToggleDarkMode={toggleDarkMode}
isMobile={isMobile}
onCloseProtectionChange={setCloseProtection}
/>
</div>
</ResizablePanel>
</ResizablePanelGroup>
</div>
);
)
}

View File

@@ -1,12 +1,12 @@
import { MetadataRoute } from 'next'
import type { MetadataRoute } from "next"
export default function robots(): MetadataRoute.Robots {
return {
rules: {
userAgent: '*',
allow: '/',
disallow: '/api/',
userAgent: "*",
allow: "/",
disallow: "/api/",
},
sitemap: 'https://next-ai-drawio.jiang.jp/sitemap.xml',
sitemap: "https://next-ai-drawio.jiang.jp/sitemap.xml",
}
}

View File

@@ -1,17 +1,17 @@
import { MetadataRoute } from 'next'
import type { MetadataRoute } from "next"
export default function sitemap(): MetadataRoute.Sitemap {
return [
{
url: 'https://next-ai-drawio.jiang.jp',
url: "https://next-ai-drawio.jiang.jp",
lastModified: new Date(),
changeFrequency: 'weekly',
changeFrequency: "weekly",
priority: 1,
},
{
url: 'https://next-ai-drawio.jiang.jp/about',
url: "https://next-ai-drawio.jiang.jp/about",
lastModified: new Date(),
changeFrequency: 'monthly',
changeFrequency: "monthly",
priority: 0.8,
},
]

83
biome.json Normal file
View File

@@ -0,0 +1,83 @@
{
"$schema": "https://biomejs.dev/schemas/2.3.8/schema.json",
"vcs": {
"enabled": true,
"clientKind": "git",
"useIgnoreFile": true
},
"files": {
"ignoreUnknown": false
},
"formatter": {
"enabled": true,
"indentStyle": "space",
"indentWidth": 4
},
"linter": {
"enabled": true,
"rules": {
"recommended": true,
"complexity": {
"noImportantStyles": "off"
},
"suspicious": {
"noExplicitAny": "off",
"noArrayIndexKey": "off",
"noImplicitAnyLet": "off",
"noAssignInExpressions": "off"
},
"a11y": {
"useButtonType": "off",
"noAutofocus": "off",
"noStaticElementInteractions": "off",
"useKeyWithClickEvents": "off",
"noLabelWithoutControl": "off",
"noNoninteractiveTabindex": "off"
},
"correctness": {
"useExhaustiveDependencies": "off"
},
"style": {
"useNodejsImportProtocol": "off",
"useTemplate": "off"
},
"security": {
"noDangerouslySetInnerHtml": "off"
}
}
},
"javascript": {
"formatter": {
"quoteStyle": "double",
"semicolons": "asNeeded"
}
},
"css": {
"parser": {
"cssModules": true,
"tailwindDirectives": true
}
},
"assist": {
"enabled": true,
"actions": {
"source": {
"organizeImports": "on"
}
}
},
"overrides": [
{
"includes": ["components/ui/**"],
"formatter": {
"enabled": false
},
"linter": {
"enabled": false
},
"assist": {
"enabled": false
}
}
]
}

View File

@@ -1,21 +1,21 @@
{
"$schema": "https://ui.shadcn.com/schema.json",
"style": "new-york",
"rsc": true,
"tsx": true,
"tailwind": {
"config": "",
"css": "app/globals.css",
"baseColor": "neutral",
"cssVariables": true,
"prefix": ""
},
"aliases": {
"components": "@/components",
"utils": "@/lib/utils",
"ui": "@/components/ui",
"lib": "@/lib",
"hooks": "@/hooks"
},
"iconLibrary": "lucide"
}
"$schema": "https://ui.shadcn.com/schema.json",
"style": "new-york",
"rsc": true,
"tsx": true,
"tailwind": {
"config": "",
"css": "app/globals.css",
"baseColor": "neutral",
"cssVariables": true,
"prefix": ""
},
"aliases": {
"components": "@/components",
"utils": "@/lib/utils",
"ui": "@/components/ui",
"lib": "@/lib",
"hooks": "@/hooks"
},
"iconLibrary": "lucide"
}

View File

@@ -0,0 +1,186 @@
"use client"
import { useControllableState } from "@radix-ui/react-use-controllable-state"
import { BrainIcon, ChevronDownIcon } from "lucide-react"
import type { ComponentProps, ReactNode } from "react"
import { createContext, memo, useContext, useEffect, useState } from "react"
import {
Collapsible,
CollapsibleContent,
CollapsibleTrigger,
} from "@/components/ui/collapsible"
import { cn } from "@/lib/utils"
import { Shimmer } from "./shimmer"
type ReasoningContextValue = {
isStreaming: boolean
isOpen: boolean
setIsOpen: (open: boolean) => void
duration: number | undefined
}
const ReasoningContext = createContext<ReasoningContextValue | null>(null)
export const useReasoning = () => {
const context = useContext(ReasoningContext)
if (!context) {
throw new Error("Reasoning components must be used within Reasoning")
}
return context
}
export type ReasoningProps = ComponentProps<typeof Collapsible> & {
isStreaming?: boolean
open?: boolean
defaultOpen?: boolean
onOpenChange?: (open: boolean) => void
duration?: number
}
const AUTO_CLOSE_DELAY = 1000
const MS_IN_S = 1000
export const Reasoning = memo(
({
className,
isStreaming = false,
open,
defaultOpen = true,
onOpenChange,
duration: durationProp,
children,
...props
}: ReasoningProps) => {
const [isOpen, setIsOpen] = useControllableState({
prop: open,
defaultProp: defaultOpen,
onChange: onOpenChange,
})
const [duration, setDuration] = useControllableState({
prop: durationProp,
defaultProp: undefined,
})
const [hasAutoClosed, setHasAutoClosed] = useState(false)
const [startTime, setStartTime] = useState<number | null>(null)
// Track duration when streaming starts and ends
useEffect(() => {
if (isStreaming) {
if (startTime === null) {
setStartTime(Date.now())
}
} else if (startTime !== null) {
setDuration(Math.ceil((Date.now() - startTime) / MS_IN_S))
setStartTime(null)
}
}, [isStreaming, startTime, setDuration])
// Auto-open when streaming starts, auto-close when streaming ends (once only)
useEffect(() => {
if (defaultOpen && !isStreaming && isOpen && !hasAutoClosed) {
// Add a small delay before closing to allow user to see the content
const timer = setTimeout(() => {
setIsOpen(false)
setHasAutoClosed(true)
}, AUTO_CLOSE_DELAY)
return () => clearTimeout(timer)
}
}, [isStreaming, isOpen, defaultOpen, setIsOpen, hasAutoClosed])
const handleOpenChange = (newOpen: boolean) => {
setIsOpen(newOpen)
}
return (
<ReasoningContext.Provider
value={{ isStreaming, isOpen, setIsOpen, duration }}
>
<Collapsible
className={cn("not-prose mb-4", className)}
onOpenChange={handleOpenChange}
open={isOpen}
{...props}
>
{children}
</Collapsible>
</ReasoningContext.Provider>
)
},
)
export type ReasoningTriggerProps = ComponentProps<
typeof CollapsibleTrigger
> & {
getThinkingMessage?: (isStreaming: boolean, duration?: number) => ReactNode
}
const defaultGetThinkingMessage = (isStreaming: boolean, duration?: number) => {
if (isStreaming || duration === 0) {
return <Shimmer duration={1}>Thinking...</Shimmer>
}
if (duration === undefined) {
return <p>Thought for a few seconds</p>
}
return <p>Thought for {duration} seconds</p>
}
export const ReasoningTrigger = memo(
({
className,
children,
getThinkingMessage = defaultGetThinkingMessage,
...props
}: ReasoningTriggerProps) => {
const { isStreaming, isOpen, duration } = useReasoning()
return (
<CollapsibleTrigger
className={cn(
"flex w-full items-center gap-2 text-muted-foreground text-sm transition-colors hover:text-foreground",
className,
)}
{...props}
>
{children ?? (
<>
<BrainIcon className="size-4" />
{getThinkingMessage(isStreaming, duration)}
<ChevronDownIcon
className={cn(
"size-4 transition-transform",
isOpen ? "rotate-180" : "rotate-0",
)}
/>
</>
)}
</CollapsibleTrigger>
)
},
)
export type ReasoningContentProps = ComponentProps<
typeof CollapsibleContent
> & {
children: string
}
export const ReasoningContent = memo(
({ className, children, ...props }: ReasoningContentProps) => (
<CollapsibleContent
className={cn(
"mt-4 text-sm",
"data-[state=closed]:fade-out-0 data-[state=closed]:slide-out-to-top-2 data-[state=open]:slide-in-from-top-2 text-muted-foreground outline-none data-[state=closed]:animate-out data-[state=open]:animate-in",
className,
)}
{...props}
>
<div className="whitespace-pre-wrap">{children}</div>
</CollapsibleContent>
),
)
Reasoning.displayName = "Reasoning"
ReasoningTrigger.displayName = "ReasoningTrigger"
ReasoningContent.displayName = "ReasoningContent"

View File

@@ -0,0 +1,64 @@
"use client"
import { motion } from "motion/react"
import {
type CSSProperties,
type ElementType,
type JSX,
memo,
useMemo,
} from "react"
import { cn } from "@/lib/utils"
export type TextShimmerProps = {
children: string
as?: ElementType
className?: string
duration?: number
spread?: number
}
const ShimmerComponent = ({
children,
as: Component = "p",
className,
duration = 2,
spread = 2,
}: TextShimmerProps) => {
const MotionComponent = motion.create(
Component as keyof JSX.IntrinsicElements,
)
const dynamicSpread = useMemo(
() => (children?.length ?? 0) * spread,
[children, spread],
)
return (
<MotionComponent
animate={{ backgroundPosition: "0% center" }}
className={cn(
"relative inline-block bg-[length:250%_100%,auto] bg-clip-text text-transparent",
"[--bg:linear-gradient(90deg,#0000_calc(50%-var(--spread)),var(--color-background),#0000_calc(50%+var(--spread)))] [background-repeat:no-repeat,padding-box]",
className,
)}
initial={{ backgroundPosition: "100% center" }}
style={
{
"--spread": `${dynamicSpread}px`,
backgroundImage:
"var(--bg), linear-gradient(var(--color-muted-foreground), var(--color-muted-foreground))",
} as CSSProperties
}
transition={{
repeat: Number.POSITIVE_INFINITY,
duration,
ease: "linear",
}}
>
{children}
</MotionComponent>
)
}
export const Shimmer = memo(ShimmerComponent)

View File

@@ -1,19 +1,19 @@
import React from "react";
import { Button, buttonVariants } from "@/components/ui/button";
import type { VariantProps } from "class-variance-authority"
import type React from "react"
import { Button, type buttonVariants } from "@/components/ui/button"
import {
Tooltip,
TooltipContent,
TooltipProvider,
TooltipTrigger,
} from "@/components/ui/tooltip";
import { type VariantProps } from "class-variance-authority";
} from "@/components/ui/tooltip"
interface ButtonWithTooltipProps
extends React.ComponentProps<"button">,
VariantProps<typeof buttonVariants> {
tooltipContent: string;
children: React.ReactNode;
asChild?: boolean;
tooltipContent: string
children: React.ReactNode
asChild?: boolean
}
export function ButtonWithTooltip({
@@ -27,8 +27,10 @@ export function ButtonWithTooltip({
<TooltipTrigger asChild>
<Button {...buttonProps}>{children}</Button>
</TooltipTrigger>
<TooltipContent>{tooltipContent}</TooltipContent>
<TooltipContent className="max-w-xs text-wrap">
{tooltipContent}
</TooltipContent>
</Tooltip>
</TooltipProvider>
);
)
}

View File

@@ -1,82 +1,180 @@
"use client"
import { Cloud, FileText, GitBranch, Palette, Zap } from "lucide-react"
interface ExampleCardProps {
icon: React.ReactNode
title: string
description: string
onClick: () => void
isNew?: boolean
}
function ExampleCard({
icon,
title,
description,
onClick,
isNew,
}: ExampleCardProps) {
return (
<button
onClick={onClick}
className={`group w-full text-left p-4 rounded-xl border bg-card hover:bg-accent/50 hover:border-primary/30 transition-all duration-200 hover:shadow-sm ${
isNew
? "border-primary/40 ring-1 ring-primary/20"
: "border-border/60"
}`}
>
<div className="flex items-start gap-3">
<div
className={`w-9 h-9 rounded-lg flex items-center justify-center shrink-0 transition-colors ${
isNew
? "bg-primary/20 group-hover:bg-primary/25"
: "bg-primary/10 group-hover:bg-primary/15"
}`}
>
{icon}
</div>
<div className="min-w-0">
<div className="flex items-center gap-2">
<h3 className="text-sm font-medium text-foreground group-hover:text-primary transition-colors">
{title}
</h3>
{isNew && (
<span className="px-1.5 py-0.5 text-[10px] font-semibold bg-primary text-primary-foreground rounded">
NEW
</span>
)}
</div>
<p className="text-xs text-muted-foreground mt-0.5 line-clamp-2">
{description}
</p>
</div>
</div>
</button>
)
}
export default function ExamplePanel({
setInput,
setFiles,
}: {
setInput: (input: string) => void;
setFiles: (files: File[]) => void;
setInput: (input: string) => void
setFiles: (files: File[]) => void
}) {
// New handler for the "Replicate this flowchart" button
const handleReplicateFlowchart = async () => {
setInput("Replicate this flowchart.");
setInput("Replicate this flowchart.")
try {
// Fetch the example image
const response = await fetch("/example.png");
const blob = await response.blob();
const file = new File([blob], "example.png", { type: "image/png" });
// Set the file to the files state
setFiles([file]);
const response = await fetch("/example.png")
const blob = await response.blob()
const file = new File([blob], "example.png", { type: "image/png" })
setFiles([file])
} catch (error) {
console.error("Error loading example image:", error);
console.error("Error loading example image:", error)
}
};
}
// Handler for the "Replicate this in aws style" button
const handleReplicateArchitecture = async () => {
setInput("Replicate this in aws style");
setInput("Replicate this in aws style")
try {
// Fetch the architecture image
const response = await fetch("/architecture.png");
const blob = await response.blob();
const response = await fetch("/architecture.png")
const blob = await response.blob()
const file = new File([blob], "architecture.png", {
type: "image/png",
});
// Set the file to the files state
setFiles([file]);
})
setFiles([file])
} catch (error) {
console.error("Error loading architecture image:", error);
console.error("Error loading architecture image:", error)
}
};
}
const handlePdfExample = async () => {
setInput("Summarize this paper as a diagram")
try {
const response = await fetch("/chain-of-thought.txt")
const blob = await response.blob()
const file = new File([blob], "chain-of-thought.txt", {
type: "text/plain",
})
setFiles([file])
} catch (error) {
console.error("Error loading text file:", error)
}
}
return (
<div className="px-4 py-2 border-t border-b border-gray-100">
<p className="text-sm text-gray-500 mb-2">
{" "}
Start a conversation to generate or modify diagrams.
</p>
<p className="text-sm text-gray-500 mb-2">
{" "}
You can also upload images to use as references.
</p>
<p className="text-sm text-gray-500 mb-2">Try these examples:</p>
<div className="flex flex-wrap gap-5">
<button
className="text-xs bg-gray-100 hover:bg-gray-200 text-gray-800 font-medium py-1 px-2 rounded"
onClick={() => setInput("Give me a **animated connector** diagram of transformer's architecture")}
>
Draw diagram with Animated Connectors
</button>
<button
className="text-xs bg-gray-100 hover:bg-gray-200 text-gray-800 font-medium py-1 px-2 rounded"
onClick={handleReplicateArchitecture}
>
Create AWS architecture
</button>
<button
className="text-xs bg-gray-100 hover:bg-gray-200 text-gray-800 font-medium py-1 px-2 rounded"
onClick={handleReplicateFlowchart}
>
Replicate flowchart
</button>
<button
className="text-xs bg-gray-100 hover:bg-gray-200 text-gray-800 font-medium py-1 px-2 rounded"
onClick={() => setInput("Draw a cat for me")}
>
Draw a cat
</button>
<div className="py-6 px-2 animate-fade-in">
{/* Welcome section */}
<div className="text-center mb-6">
<h2 className="text-lg font-semibold text-foreground mb-2">
Create diagrams with AI
</h2>
<p className="text-sm text-muted-foreground max-w-xs mx-auto">
Describe what you want to create or upload an image to
replicate
</p>
</div>
{/* Examples grid */}
<div className="space-y-3">
<p className="text-xs font-medium text-muted-foreground uppercase tracking-wider px-1">
Quick Examples
</p>
<div className="grid gap-2">
<ExampleCard
icon={<FileText className="w-4 h-4 text-primary" />}
title="Paper to Diagram"
description="Upload .pdf, .txt, .md, .json, .csv, .py, .js, .ts and more"
onClick={handlePdfExample}
isNew
/>
<ExampleCard
icon={<Zap className="w-4 h-4 text-primary" />}
title="Animated Diagram"
description="Draw a transformer architecture with animated connectors"
onClick={() => {
setInput(
"Give me a **animated connector** diagram of transformer's architecture",
)
setFiles([])
}}
/>
<ExampleCard
icon={<Cloud className="w-4 h-4 text-primary" />}
title="AWS Architecture"
description="Create a cloud architecture diagram with AWS icons"
onClick={handleReplicateArchitecture}
/>
<ExampleCard
icon={<GitBranch className="w-4 h-4 text-primary" />}
title="Replicate Flowchart"
description="Upload and replicate an existing flowchart"
onClick={handleReplicateFlowchart}
/>
<ExampleCard
icon={<Palette className="w-4 h-4 text-primary" />}
title="Creative Drawing"
description="Draw something fun and creative"
onClick={() => {
setInput("Draw a cat for me")
setFiles([])
}}
/>
</div>
<p className="text-[11px] text-muted-foreground/60 text-center mt-4">
Examples are cached for instant response
</p>
</div>
</div>
);
)
}

View File

@@ -1,31 +1,142 @@
"use client";
"use client"
import React, { useCallback, useRef, useEffect, useState } from "react";
import { Button } from "@/components/ui/button";
import { Textarea } from "@/components/ui/textarea";
import { ResetWarningModal } from "@/components/reset-warning-modal";
import {
Download,
History,
Image as ImageIcon,
Loader2,
Send,
RotateCcw,
Image as ImageIcon,
History,
} from "lucide-react";
import { ButtonWithTooltip } from "@/components/button-with-tooltip";
import { FilePreviewList } from "./file-preview-list";
import { useDiagram } from "@/contexts/diagram-context";
import { HistoryDialog } from "@/components/history-dialog";
Trash2,
} from "lucide-react"
import type React from "react"
import { useCallback, useEffect, useRef, useState } from "react"
import { toast } from "sonner"
import { ButtonWithTooltip } from "@/components/button-with-tooltip"
import { ErrorToast } from "@/components/error-toast"
import { HistoryDialog } from "@/components/history-dialog"
import { ResetWarningModal } from "@/components/reset-warning-modal"
import { SaveDialog } from "@/components/save-dialog"
import { Button } from "@/components/ui/button"
import { Switch } from "@/components/ui/switch"
import { Textarea } from "@/components/ui/textarea"
import {
Tooltip,
TooltipContent,
TooltipTrigger,
} from "@/components/ui/tooltip"
import { useDiagram } from "@/contexts/diagram-context"
import { isPdfFile, isTextFile } from "@/lib/pdf-utils"
import { FilePreviewList } from "./file-preview-list"
const MAX_IMAGE_SIZE = 2 * 1024 * 1024 // 2MB
const MAX_FILES = 5
function isValidFileType(file: File): boolean {
return file.type.startsWith("image/") || isPdfFile(file) || isTextFile(file)
}
function formatFileSize(bytes: number): string {
const mb = bytes / 1024 / 1024
if (mb < 0.01) return `${(bytes / 1024).toFixed(0)}KB`
return `${mb.toFixed(2)}MB`
}
function showErrorToast(message: React.ReactNode) {
toast.custom(
(t) => (
<ErrorToast message={message} onDismiss={() => toast.dismiss(t)} />
),
{ duration: 5000 },
)
}
interface ValidationResult {
validFiles: File[]
errors: string[]
}
function validateFiles(
newFiles: File[],
existingCount: number,
): ValidationResult {
const errors: string[] = []
const validFiles: File[] = []
const availableSlots = MAX_FILES - existingCount
if (availableSlots <= 0) {
errors.push(`Maximum ${MAX_FILES} files allowed`)
return { validFiles, errors }
}
for (const file of newFiles) {
if (validFiles.length >= availableSlots) {
errors.push(`Only ${availableSlots} more file(s) allowed`)
break
}
if (!isValidFileType(file)) {
errors.push(`"${file.name}" is not a supported file type`)
continue
}
// Only check size for images (PDFs/text files are extracted client-side, so file size doesn't matter)
const isExtractedFile = isPdfFile(file) || isTextFile(file)
if (!isExtractedFile && file.size > MAX_IMAGE_SIZE) {
const maxSizeMB = MAX_IMAGE_SIZE / 1024 / 1024
errors.push(
`"${file.name}" is ${formatFileSize(file.size)} (exceeds ${maxSizeMB}MB)`,
)
} else {
validFiles.push(file)
}
}
return { validFiles, errors }
}
function showValidationErrors(errors: string[]) {
if (errors.length === 0) return
if (errors.length === 1) {
showErrorToast(
<span className="text-muted-foreground">{errors[0]}</span>,
)
} else {
showErrorToast(
<div className="flex flex-col gap-1">
<span className="font-medium">
{errors.length} files rejected:
</span>
<ul className="text-muted-foreground text-xs list-disc list-inside">
{errors.slice(0, 3).map((err) => (
<li key={err}>{err}</li>
))}
{errors.length > 3 && (
<li>...and {errors.length - 3} more</li>
)}
</ul>
</div>,
)
}
}
interface ChatInputProps {
input: string;
status: "submitted" | "streaming" | "ready" | "error";
onSubmit: (e: React.FormEvent<HTMLFormElement>) => void;
onChange: (e: React.ChangeEvent<HTMLTextAreaElement>) => void;
onClearChat: () => void;
files?: File[];
onFileChange?: (files: File[]) => void;
showHistory?: boolean;
onToggleHistory?: (show: boolean) => void;
input: string
status: "submitted" | "streaming" | "ready" | "error"
onSubmit: (e: React.FormEvent<HTMLFormElement>) => void
onChange: (e: React.ChangeEvent<HTMLTextAreaElement>) => void
onClearChat: () => void
files?: File[]
onFileChange?: (files: File[]) => void
pdfData?: Map<
File,
{ text: string; charCount: number; isExtracting: boolean }
>
showHistory?: boolean
onToggleHistory?: (show: boolean) => void
sessionId?: string
error?: Error | null
minimalStyle?: boolean
onMinimalStyleChange?: (value: boolean) => void
}
export function ChatInput({
@@ -36,245 +147,322 @@ export function ChatInput({
onClearChat,
files = [],
onFileChange = () => {},
pdfData = new Map(),
showHistory = false,
onToggleHistory = () => {},
sessionId,
error = null,
minimalStyle = false,
onMinimalStyleChange = () => {},
}: ChatInputProps) {
const { diagramHistory } = useDiagram();
const textareaRef = useRef<HTMLTextAreaElement>(null);
const fileInputRef = useRef<HTMLInputElement>(null);
const [isDragging, setIsDragging] = useState(false);
const [showClearDialog, setShowClearDialog] = useState(false);
const { diagramHistory, saveDiagramToFile } = useDiagram()
const textareaRef = useRef<HTMLTextAreaElement>(null)
const fileInputRef = useRef<HTMLInputElement>(null)
const [isDragging, setIsDragging] = useState(false)
const [showClearDialog, setShowClearDialog] = useState(false)
const [showSaveDialog, setShowSaveDialog] = useState(false)
// Debug: Log status changes
const isDisabled = status === "streaming" || status === "submitted";
useEffect(() => {
console.log('[ChatInput] Status changed to:', status, '| Input disabled:', isDisabled);
}, [status, isDisabled]);
// Allow retry when there's an error (even if status is still "streaming" or "submitted")
const isDisabled =
(status === "streaming" || status === "submitted") && !error
// Auto-resize textarea based on content
const adjustTextareaHeight = useCallback(() => {
const textarea = textareaRef.current;
const textarea = textareaRef.current
if (textarea) {
textarea.style.height = "auto";
textarea.style.height = `${Math.min(textarea.scrollHeight, 200)}px`;
textarea.style.height = "auto"
textarea.style.height = `${Math.min(textarea.scrollHeight, 200)}px`
}
}, []);
}, [])
// Handle programmatic input changes (e.g., setInput("") after form submission)
useEffect(() => {
adjustTextareaHeight();
}, [input, adjustTextareaHeight]);
adjustTextareaHeight()
}, [input, adjustTextareaHeight])
const handleChange = (e: React.ChangeEvent<HTMLTextAreaElement>) => {
onChange(e)
adjustTextareaHeight()
}
// Handle keyboard shortcuts and paste events
const handleKeyDown = (e: React.KeyboardEvent) => {
if ((e.metaKey || e.ctrlKey) && e.key === "Enter") {
e.preventDefault();
const form = e.currentTarget.closest("form");
e.preventDefault()
const form = e.currentTarget.closest("form")
if (form && input.trim() && !isDisabled) {
form.requestSubmit();
form.requestSubmit()
}
}
};
}
// Handle clipboard paste
const handlePaste = async (e: React.ClipboardEvent) => {
if (isDisabled) return;
if (isDisabled) return
const items = e.clipboardData.items;
const items = e.clipboardData.items
const imageItems = Array.from(items).filter((item) =>
item.type.startsWith("image/")
);
item.type.startsWith("image/"),
)
if (imageItems.length > 0) {
const imageFiles = await Promise.all(
imageItems.map(async (item) => {
const file = item.getAsFile();
if (!file) return null;
// Create a new file with a unique name
return new File(
[file],
`pasted-image-${Date.now()}.${file.type.split("/")[1]}`,
{
type: file.type,
}
);
})
);
const imageFiles = (
await Promise.all(
imageItems.map(async (item, index) => {
const file = item.getAsFile()
if (!file) return null
return new File(
[file],
`pasted-image-${Date.now()}-${index}.${file.type.split("/")[1]}`,
{ type: file.type },
)
}),
)
).filter((f): f is File => f !== null)
const validFiles = imageFiles.filter(
(file): file is File => file !== null
);
const { validFiles, errors } = validateFiles(
imageFiles,
files.length,
)
showValidationErrors(errors)
if (validFiles.length > 0) {
onFileChange([...files, ...validFiles]);
onFileChange([...files, ...validFiles])
}
}
};
}
// Handle file changes
const handleFileChange = (e: React.ChangeEvent<HTMLInputElement>) => {
const newFiles = Array.from(e.target.files || []);
onFileChange([...files, ...newFiles]);
};
// Remove individual file
const handleRemoveFile = (fileToRemove: File) => {
onFileChange(files.filter((file) => file !== fileToRemove));
if (fileInputRef.current) {
fileInputRef.current.value = "";
const newFiles = Array.from(e.target.files || [])
const { validFiles, errors } = validateFiles(newFiles, files.length)
showValidationErrors(errors)
if (validFiles.length > 0) {
onFileChange([...files, ...validFiles])
}
};
// Reset input so same file can be selected again
if (fileInputRef.current) {
fileInputRef.current.value = ""
}
}
const handleRemoveFile = (fileToRemove: File) => {
onFileChange(files.filter((file) => file !== fileToRemove))
if (fileInputRef.current) {
fileInputRef.current.value = ""
}
}
// Trigger file input click
const triggerFileInput = () => {
fileInputRef.current?.click();
};
fileInputRef.current?.click()
}
// Handle drag events
const handleDragOver = (e: React.DragEvent<HTMLFormElement>) => {
e.preventDefault();
e.stopPropagation();
setIsDragging(true);
};
e.preventDefault()
e.stopPropagation()
setIsDragging(true)
}
const handleDragLeave = (e: React.DragEvent<HTMLFormElement>) => {
e.preventDefault();
e.stopPropagation();
setIsDragging(false);
};
e.preventDefault()
e.stopPropagation()
setIsDragging(false)
}
const handleDrop = (e: React.DragEvent<HTMLFormElement>) => {
e.preventDefault();
e.stopPropagation();
setIsDragging(false);
e.preventDefault()
e.stopPropagation()
setIsDragging(false)
if (isDisabled) return;
if (isDisabled) return
const droppedFiles = e.dataTransfer.files;
const droppedFiles = e.dataTransfer.files
const supportedFiles = Array.from(droppedFiles).filter((file) =>
isValidFileType(file),
)
// Only process image files
const imageFiles = Array.from(droppedFiles).filter((file) =>
file.type.startsWith("image/")
);
if (imageFiles.length > 0) {
onFileChange([...files, ...imageFiles]);
const { validFiles, errors } = validateFiles(
supportedFiles,
files.length,
)
showValidationErrors(errors)
if (validFiles.length > 0) {
onFileChange([...files, ...validFiles])
}
};
}
// Handle clearing conversation and diagram
const handleClear = () => {
onClearChat();
setShowClearDialog(false);
};
onClearChat()
setShowClearDialog(false)
}
return (
<form
onSubmit={onSubmit}
className={`w-full space-y-2 ${
className={`w-full transition-all duration-200 ${
isDragging
? "border-2 border-dashed border-primary p-4 rounded-lg bg-muted/20"
? "ring-2 ring-primary ring-offset-2 rounded-2xl"
: ""
}`}
onDragOver={handleDragOver}
onDragLeave={handleDragLeave}
onDrop={handleDrop}
>
<FilePreviewList files={files} onRemoveFile={handleRemoveFile} />
<Textarea
ref={textareaRef}
value={input}
onChange={onChange}
onKeyDown={handleKeyDown}
onPaste={handlePaste}
placeholder="Describe what changes you want to make to the diagram
or upload(paste) an image to replicate a diagram.
(Press Cmd/Ctrl + Enter to send)"
disabled={isDisabled}
aria-label="Chat input"
className="min-h-[80px] resize-none transition-all duration-200 px-1 py-0"
/>
<div className="flex items-center gap-2">
<div className="mr-auto">
<ButtonWithTooltip
type="button"
variant="ghost"
size="icon"
onClick={() => setShowClearDialog(true)}
tooltipContent="Clear current conversation and diagram"
>
<RotateCcw className="mr-2 h-4 w-4" />
</ButtonWithTooltip>
{/* Warning Modal */}
<ResetWarningModal
open={showClearDialog}
onOpenChange={setShowClearDialog}
onClear={handleClear}
/>
<HistoryDialog
showHistory={showHistory}
onToggleHistory={onToggleHistory}
{/* File previews */}
{files.length > 0 && (
<div className="mb-3">
<FilePreviewList
files={files}
onRemoveFile={handleRemoveFile}
pdfData={pdfData}
/>
</div>
<div className="flex gap-2">
{/* History Button */}
<ButtonWithTooltip
type="button"
variant="outline"
size="icon"
onClick={() => onToggleHistory(true)}
disabled={
isDisabled ||
diagramHistory.length === 0
}
title="Diagram History"
tooltipContent="View diagram history"
>
<History className="h-4 w-4" />
</ButtonWithTooltip>
)}
<Button
type="button"
variant="outline"
size="icon"
onClick={triggerFileInput}
disabled={isDisabled}
title="Upload image"
>
<ImageIcon className="h-4 w-4" />
</Button>
{/* Input container */}
<div className="relative rounded-2xl border border-border bg-background shadow-sm focus-within:ring-2 focus-within:ring-primary/20 focus-within:border-primary/50 transition-all duration-200">
<Textarea
ref={textareaRef}
value={input}
onChange={handleChange}
onKeyDown={handleKeyDown}
onPaste={handlePaste}
placeholder="Describe your diagram or upload a file..."
disabled={isDisabled}
aria-label="Chat input"
className="min-h-[60px] max-h-[200px] resize-none border-0 bg-transparent px-4 py-3 text-sm focus-visible:ring-0 focus-visible:ring-offset-0 placeholder:text-muted-foreground/60"
/>
<input
type="file"
ref={fileInputRef}
className="hidden"
onChange={handleFileChange}
accept="image/*"
multiple
disabled={isDisabled}
/>
{/* Action bar */}
<div className="flex items-center justify-between px-3 py-2 border-t border-border/50">
{/* Left actions */}
<div className="flex items-center gap-1">
<ButtonWithTooltip
type="button"
variant="ghost"
size="sm"
onClick={() => setShowClearDialog(true)}
tooltipContent="Clear conversation"
className="h-8 w-8 p-0 text-muted-foreground hover:text-destructive hover:bg-destructive/10"
>
<Trash2 className="h-4 w-4" />
</ButtonWithTooltip>
<ResetWarningModal
open={showClearDialog}
onOpenChange={setShowClearDialog}
onClear={handleClear}
/>
<HistoryDialog
showHistory={showHistory}
onToggleHistory={onToggleHistory}
/>
<Tooltip>
<TooltipTrigger asChild>
<div className="flex items-center gap-1.5">
<Switch
id="minimal-style"
checked={minimalStyle}
onCheckedChange={onMinimalStyleChange}
className="scale-75"
/>
<label
htmlFor="minimal-style"
className={`text-xs cursor-pointer select-none ${
minimalStyle
? "text-primary font-medium"
: "text-muted-foreground"
}`}
>
{minimalStyle ? "Minimal" : "Styled"}
</label>
</div>
</TooltipTrigger>
<TooltipContent side="top">
Use minimal for faster generation (no colors)
</TooltipContent>
</Tooltip>
</div>
{/* Right actions */}
<div className="flex items-center gap-1">
<ButtonWithTooltip
type="button"
variant="ghost"
size="sm"
onClick={() => onToggleHistory(true)}
disabled={isDisabled || diagramHistory.length === 0}
tooltipContent="Diagram history"
className="h-8 w-8 p-0 text-muted-foreground hover:text-foreground"
>
<History className="h-4 w-4" />
</ButtonWithTooltip>
<ButtonWithTooltip
type="button"
variant="ghost"
size="sm"
onClick={() => setShowSaveDialog(true)}
disabled={isDisabled}
tooltipContent="Save diagram"
className="h-8 w-8 p-0 text-muted-foreground hover:text-foreground"
>
<Download className="h-4 w-4" />
</ButtonWithTooltip>
<SaveDialog
open={showSaveDialog}
onOpenChange={setShowSaveDialog}
onSave={(filename, format) =>
saveDiagramToFile(filename, format, sessionId)
}
defaultFilename={`diagram-${new Date()
.toISOString()
.slice(0, 10)}`}
/>
<ButtonWithTooltip
type="button"
variant="ghost"
size="sm"
onClick={triggerFileInput}
disabled={isDisabled}
tooltipContent="Upload file (image, PDF, text)"
className="h-8 w-8 p-0 text-muted-foreground hover:text-foreground"
>
<ImageIcon className="h-4 w-4" />
</ButtonWithTooltip>
<input
type="file"
ref={fileInputRef}
className="hidden"
onChange={handleFileChange}
accept="image/*,.pdf,application/pdf,text/*,.md,.markdown,.json,.csv,.xml,.yaml,.yml,.toml"
multiple
disabled={isDisabled}
/>
<div className="w-px h-5 bg-border mx-1" />
<Button
type="submit"
disabled={isDisabled || !input.trim()}
size="sm"
className="h-8 px-4 rounded-xl font-medium shadow-sm"
aria-label={
isDisabled ? "Sending..." : "Send message"
}
>
{isDisabled ? (
<Loader2 className="h-4 w-4 animate-spin" />
) : (
<>
<Send className="h-4 w-4 mr-1.5" />
Send
</>
)}
</Button>
</div>
</div>
<Button
type="submit"
disabled={isDisabled || !input.trim()}
className="transition-opacity"
aria-label={
isDisabled
? "Sending message..."
: "Send message"
}
>
{isDisabled ? (
<Loader2 className="mr-2 h-4 w-4 animate-spin" />
) : (
<Send className="mr-2 h-4 w-4" />
)}
Send
</Button>
</div>
</form>
);
)
}

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

53
components/code-block.tsx Normal file
View File

@@ -0,0 +1,53 @@
"use client"
import { Highlight, themes } from "prism-react-renderer"
interface CodeBlockProps {
code: string
language?: "xml" | "json"
}
export function CodeBlock({ code, language = "xml" }: CodeBlockProps) {
return (
<div className="overflow-hidden w-full">
<Highlight theme={themes.github} code={code} language={language}>
{({
className: _className,
style,
tokens,
getLineProps,
getTokenProps,
}) => (
<pre
className="text-[11px] leading-relaxed overflow-x-auto overflow-y-auto max-h-48 scrollbar-thin break-all"
style={{
...style,
fontFamily:
"var(--font-mono), ui-monospace, monospace",
backgroundColor: "transparent",
margin: 0,
padding: 0,
wordBreak: "break-all",
whiteSpace: "pre-wrap",
}}
>
{tokens.map((line, i) => (
<div
key={i}
{...getLineProps({ line })}
style={{ wordBreak: "break-all" }}
>
{line.map((token, key) => (
<span
key={key}
{...getTokenProps({ token })}
/>
))}
</div>
))}
</pre>
)}
</Highlight>
</div>
)
}

View File

@@ -0,0 +1,44 @@
"use client"
import type React from "react"
interface ErrorToastProps {
message: React.ReactNode
onDismiss: () => void
}
export function ErrorToast({ message, onDismiss }: ErrorToastProps) {
const handleKeyDown = (e: React.KeyboardEvent) => {
if (e.key === "Enter" || e.key === " " || e.key === "Escape") {
e.preventDefault()
onDismiss()
}
}
return (
<div
role="alert"
aria-live="polite"
tabIndex={0}
onClick={onDismiss}
onKeyDown={handleKeyDown}
className="flex items-center gap-3 bg-card border border-border/50 px-4 py-3 rounded-xl shadow-sm cursor-pointer hover:bg-muted/50 focus:outline-none focus:ring-2 focus:ring-primary/50 transition-colors"
>
<div className="flex items-center justify-center w-8 h-8 rounded-full bg-destructive/10 flex-shrink-0">
<svg
className="w-4 h-4 text-destructive"
viewBox="0 0 20 20"
fill="currentColor"
aria-hidden="true"
>
<path
fillRule="evenodd"
d="M4.293 4.293a1 1 0 011.414 0L10 8.586l4.293-4.293a1 1 0 111.414 1.414L11.414 10l4.293 4.293a1 1 0 01-1.414 1.414L10 11.414l-4.293 4.293a1 1 0 01-1.414-1.414L8.586 10 4.293 5.707a1 1 0 010-1.414z"
clipRule="evenodd"
/>
</svg>
</div>
<span className="text-sm text-foreground">{message}</span>
</div>
)
}

View File

@@ -1,49 +1,142 @@
"use client";
"use client"
import React, { useEffect, useState } from "react";
import Image from "next/image";
import { X } from "lucide-react";
import { FileCode, FileText, Loader2, X } from "lucide-react"
import Image from "next/image"
import { useEffect, useRef, useState } from "react"
import { isPdfFile, isTextFile } from "@/lib/pdf-utils"
interface FilePreviewListProps {
files: File[];
onRemoveFile: (fileToRemove: File) => void;
function formatCharCount(count: number): string {
if (count >= 1000) {
return `${(count / 1000).toFixed(1)}k`
}
return String(count)
}
export function FilePreviewList({ files, onRemoveFile }: FilePreviewListProps) {
const [selectedImage, setSelectedImage] = useState<string | null>(null);
interface FilePreviewListProps {
files: File[]
onRemoveFile: (fileToRemove: File) => void
pdfData?: Map<
File,
{ text: string; charCount: number; isExtracting: boolean }
>
}
// Cleanup object URLs on unmount
export function FilePreviewList({
files,
onRemoveFile,
pdfData = new Map(),
}: FilePreviewListProps) {
const [selectedImage, setSelectedImage] = useState<string | null>(null)
const [imageUrls, setImageUrls] = useState<Map<File, string>>(new Map())
const imageUrlsRef = useRef<Map<File, string>>(new Map())
// Create and cleanup object URLs when files change
useEffect(() => {
const objectUrls = files
.filter((file) => file.type.startsWith("image/"))
.map((file) => URL.createObjectURL(file));
const currentUrls = imageUrlsRef.current
const newUrls = new Map<File, string>()
files.forEach((file) => {
if (file.type.startsWith("image/")) {
// Reuse existing URL if file is already tracked
const existingUrl = currentUrls.get(file)
if (existingUrl) {
newUrls.set(file, existingUrl)
} else {
newUrls.set(file, URL.createObjectURL(file))
}
}
})
// Revoke URLs for files that are no longer in the list
currentUrls.forEach((url, file) => {
if (!newUrls.has(file)) {
URL.revokeObjectURL(url)
}
})
imageUrlsRef.current = newUrls
setImageUrls(newUrls)
}, [files])
// Cleanup all URLs on unmount only
useEffect(() => {
return () => {
objectUrls.forEach(URL.revokeObjectURL);
};
}, [files]);
imageUrlsRef.current.forEach((url) => {
URL.revokeObjectURL(url)
})
// Clear the ref so StrictMode remount creates fresh URLs
imageUrlsRef.current = new Map()
}
}, [])
if (files.length === 0) return null;
// Clear selected image if its URL was revoked
useEffect(() => {
if (
selectedImage &&
!Array.from(imageUrls.values()).includes(selectedImage)
) {
setSelectedImage(null)
}
}, [imageUrls, selectedImage])
if (files.length === 0) return null
return (
<>
<div className="flex flex-wrap gap-2 mt-2 p-2 bg-muted/50 rounded-md">
{files.map((file, index) => {
const imageUrl = file.type.startsWith("image/") ? URL.createObjectURL(file) : null;
const imageUrl = imageUrls.get(file) || null
const pdfInfo = pdfData.get(file)
return (
<div key={file.name + index} className="relative group">
<div
className="w-20 h-20 border rounded-md overflow-hidden bg-muted cursor-pointer"
onClick={() => imageUrl && setSelectedImage(imageUrl)}
className={`w-20 h-20 border rounded-md overflow-hidden bg-muted ${
file.type.startsWith("image/") && imageUrl
? "cursor-pointer"
: ""
}`}
onClick={() =>
file.type.startsWith("image/") &&
imageUrl &&
setSelectedImage(imageUrl)
}
>
{file.type.startsWith("image/") ? (
{file.type.startsWith("image/") && imageUrl ? (
<Image
src={imageUrl!}
src={imageUrl}
alt={file.name}
width={80}
height={80}
className="object-cover w-full h-full"
unoptimized
/>
) : isPdfFile(file) || isTextFile(file) ? (
<div className="flex flex-col items-center justify-center h-full p-1">
{pdfInfo?.isExtracting ? (
<Loader2 className="h-6 w-6 text-blue-500 mb-1 animate-spin" />
) : isPdfFile(file) ? (
<FileText className="h-6 w-6 text-red-500 mb-1" />
) : (
<FileCode className="h-6 w-6 text-blue-500 mb-1" />
)}
<span className="text-xs text-center truncate w-full px-1">
{file.name.length > 10
? `${file.name.slice(0, 7)}...`
: file.name}
</span>
{pdfInfo?.isExtracting ? (
<span className="text-[10px] text-muted-foreground">
Reading...
</span>
) : pdfInfo?.charCount ? (
<span className="text-[10px] text-green-600 font-medium">
{formatCharCount(
pdfInfo.charCount,
)}{" "}
chars
</span>
) : null}
</div>
) : (
<div className="flex items-center justify-center h-full text-xs text-center p-1">
{file.name}
@@ -59,7 +152,7 @@ export function FilePreviewList({ files, onRemoveFile }: FilePreviewListProps) {
<X className="h-3 w-3" />
</button>
</div>
);
)
})}
</div>
@@ -84,10 +177,11 @@ export function FilePreviewList({ files, onRemoveFile }: FilePreviewListProps) {
height={900}
className="object-contain max-w-full max-h-[90vh] w-auto h-auto"
onClick={(e) => e.stopPropagation()}
unoptimized
/>
</div>
</div>
)}
</>
);
)
}

View File

@@ -1,5 +1,8 @@
"use client";
"use client"
import Image from "next/image"
import { useState } from "react"
import { Button } from "@/components/ui/button"
import {
Dialog,
DialogContent,
@@ -7,21 +10,33 @@ import {
DialogFooter,
DialogHeader,
DialogTitle,
} from "@/components/ui/dialog";
import { Button } from "@/components/ui/button";
import Image from "next/image";
import { useDiagram } from "@/contexts/diagram-context";
} from "@/components/ui/dialog"
import { useDiagram } from "@/contexts/diagram-context"
interface HistoryDialogProps {
showHistory: boolean;
onToggleHistory: (show: boolean) => void;
showHistory: boolean
onToggleHistory: (show: boolean) => void
}
export function HistoryDialog({
showHistory,
onToggleHistory,
}: HistoryDialogProps) {
const { loadDiagram: onDisplayChart, diagramHistory } = useDiagram();
const { loadDiagram: onDisplayChart, diagramHistory } = useDiagram()
const [selectedIndex, setSelectedIndex] = useState<number | null>(null)
const handleClose = () => {
setSelectedIndex(null)
onToggleHistory(false)
}
const handleConfirmRestore = () => {
if (selectedIndex !== null) {
// Skip validation for trusted history snapshots
onDisplayChart(diagramHistory[selectedIndex].xml, true)
handleClose()
}
}
return (
<Dialog open={showHistory} onOpenChange={onToggleHistory}>
@@ -45,11 +60,12 @@ export function HistoryDialog({
{diagramHistory.map((item, index) => (
<div
key={index}
className="border rounded-md p-2 cursor-pointer hover:border-primary transition-colors"
onClick={() => {
onDisplayChart(item.xml);
onToggleHistory(false);
}}
className={`border rounded-md p-2 cursor-pointer hover:border-primary transition-colors ${
selectedIndex === index
? "border-primary ring-2 ring-primary"
: ""
}`}
onClick={() => setSelectedIndex(index)}
>
<div className="aspect-video bg-white rounded overflow-hidden flex items-center justify-center">
<Image
@@ -69,14 +85,28 @@ export function HistoryDialog({
)}
<DialogFooter>
<Button
variant="outline"
onClick={() => onToggleHistory(false)}
>
Close
</Button>
{selectedIndex !== null ? (
<>
<div className="flex-1 text-sm text-muted-foreground">
Restore to Version {selectedIndex + 1}?
</div>
<Button
variant="outline"
onClick={() => setSelectedIndex(null)}
>
Cancel
</Button>
<Button onClick={handleConfirmRestore}>
Confirm
</Button>
</>
) : (
<Button variant="outline" onClick={handleClose}>
Close
</Button>
)}
</DialogFooter>
</DialogContent>
</Dialog>
);
)
}

View File

@@ -0,0 +1,115 @@
"use client"
import { Coffee, X } from "lucide-react"
import Link from "next/link"
import type React from "react"
import { FaGithub } from "react-icons/fa"
interface QuotaLimitToastProps {
type?: "request" | "token"
used: number
limit: number
onDismiss: () => void
}
export function QuotaLimitToast({
type = "request",
used,
limit,
onDismiss,
}: QuotaLimitToastProps) {
const isTokenLimit = type === "token"
const formatNumber = (n: number) =>
n >= 1000 ? `${(n / 1000).toFixed(1)}k` : n.toString()
const handleKeyDown = (e: React.KeyboardEvent) => {
if (e.key === "Escape") {
e.preventDefault()
onDismiss()
}
}
return (
<div
role="alert"
aria-live="polite"
tabIndex={0}
onKeyDown={handleKeyDown}
className="relative w-[400px] overflow-hidden rounded-xl border border-border/50 bg-card p-5 shadow-soft animate-message-in"
>
{/* Close button */}
<button
onClick={onDismiss}
className="absolute right-3 top-3 p-1.5 rounded-full text-muted-foreground/60 hover:text-foreground hover:bg-muted transition-colors"
aria-label="Dismiss"
>
<X className="w-4 h-4" />
</button>
{/* Title row with icon */}
<div className="flex items-center gap-2.5 mb-3 pr-6">
<div className="flex-shrink-0 w-8 h-8 rounded-lg bg-accent flex items-center justify-center">
<Coffee
className="w-4 h-4 text-accent-foreground"
strokeWidth={2}
/>
</div>
<h3 className="font-semibold text-foreground text-sm">
{isTokenLimit
? "Daily Token Limit Reached"
: "Daily Quota Reached"}
</h3>
<span className="px-2 py-0.5 text-xs font-medium rounded-md bg-muted text-muted-foreground">
{isTokenLimit
? `${formatNumber(used)}/${formatNumber(limit)} tokens`
: `${used}/${limit}`}
</span>
</div>
{/* Message */}
<div className="text-sm text-muted-foreground leading-relaxed mb-4 space-y-2">
<p>
Oops you've reached the daily{" "}
{isTokenLimit ? "token" : "API"} limit for this demo! As an
indie developer covering all the API costs myself, I have to
set these limits to keep things sustainable.{" "}
<Link
href="/about"
target="_blank"
rel="noopener noreferrer"
className="inline-flex items-center gap-1 text-amber-600 font-medium hover:text-amber-700 hover:underline"
>
Learn more
</Link>
</p>
<p>
<strong>Tip:</strong> You can use your own API key (click
the Settings icon) or self-host the project to bypass these
limits.
</p>
<p>Your limit resets tomorrow. Thanks for understanding!</p>
</div>
{/* Action buttons */}
<div className="flex items-center gap-2">
<a
href="https://github.com/DayuanJiang/next-ai-draw-io"
target="_blank"
rel="noopener noreferrer"
className="inline-flex items-center gap-1.5 px-3 py-1.5 text-xs font-medium rounded-lg bg-primary text-primary-foreground hover:bg-primary/90 transition-colors"
>
<FaGithub className="w-3.5 h-3.5" />
Self-host
</a>
<a
href="https://github.com/sponsors/DayuanJiang"
target="_blank"
rel="noopener noreferrer"
className="inline-flex items-center gap-1.5 px-3 py-1.5 text-xs font-medium rounded-lg border border-border text-foreground hover:bg-muted transition-colors"
>
<Coffee className="w-3.5 h-3.5" />
Sponsor
</a>
</div>
</div>
)
}

View File

@@ -1,6 +1,6 @@
"use client";
"use client"
import { Button } from "@/components/ui/button";
import { Button } from "@/components/ui/button"
import {
Dialog,
DialogContent,
@@ -8,12 +8,12 @@ import {
DialogFooter,
DialogHeader,
DialogTitle,
} from "@/components/ui/dialog";
} from "@/components/ui/dialog"
interface ResetWarningModalProps {
open: boolean;
onOpenChange: (open: boolean) => void;
onClear: () => void;
open: boolean
onOpenChange: (open: boolean) => void
onClear: () => void
}
export function ResetWarningModal({
@@ -44,5 +44,5 @@ export function ResetWarningModal({
</DialogFooter>
</DialogContent>
</Dialog>
);
)
}

128
components/save-dialog.tsx Normal file
View File

@@ -0,0 +1,128 @@
"use client"
import { useEffect, useState } from "react"
import { Button } from "@/components/ui/button"
import {
Dialog,
DialogContent,
DialogFooter,
DialogHeader,
DialogTitle,
} from "@/components/ui/dialog"
import { Input } from "@/components/ui/input"
import {
Select,
SelectContent,
SelectItem,
SelectTrigger,
SelectValue,
} from "@/components/ui/select"
export type ExportFormat = "drawio" | "png" | "svg"
const FORMAT_OPTIONS: {
value: ExportFormat
label: string
extension: string
}[] = [
{ value: "drawio", label: "Draw.io XML", extension: ".drawio" },
{ value: "png", label: "PNG Image", extension: ".png" },
{ value: "svg", label: "SVG Image", extension: ".svg" },
]
interface SaveDialogProps {
open: boolean
onOpenChange: (open: boolean) => void
onSave: (filename: string, format: ExportFormat) => void
defaultFilename: string
}
export function SaveDialog({
open,
onOpenChange,
onSave,
defaultFilename,
}: SaveDialogProps) {
const [filename, setFilename] = useState(defaultFilename)
const [format, setFormat] = useState<ExportFormat>("drawio")
useEffect(() => {
if (open) {
setFilename(defaultFilename)
}
}, [open, defaultFilename])
const handleSave = () => {
const finalFilename = filename.trim() || defaultFilename
onSave(finalFilename, format)
onOpenChange(false)
}
const handleKeyDown = (e: React.KeyboardEvent) => {
if (e.key === "Enter") {
e.preventDefault()
handleSave()
}
}
const currentFormat = FORMAT_OPTIONS.find((f) => f.value === format)
return (
<Dialog open={open} onOpenChange={onOpenChange}>
<DialogContent className="sm:max-w-md">
<DialogHeader>
<DialogTitle>Save Diagram</DialogTitle>
</DialogHeader>
<div className="space-y-4">
<div className="space-y-2">
<label className="text-sm font-medium">Format</label>
<Select
value={format}
onValueChange={(v) => setFormat(v as ExportFormat)}
>
<SelectTrigger>
<SelectValue />
</SelectTrigger>
<SelectContent>
{FORMAT_OPTIONS.map((opt) => (
<SelectItem
key={opt.value}
value={opt.value}
>
{opt.label}
</SelectItem>
))}
</SelectContent>
</Select>
</div>
<div className="space-y-2">
<label className="text-sm font-medium">Filename</label>
<div className="flex items-stretch">
<Input
value={filename}
onChange={(e) => setFilename(e.target.value)}
onKeyDown={handleKeyDown}
placeholder="Enter filename"
autoFocus
onFocus={(e) => e.target.select()}
className="rounded-r-none border-r-0 focus-visible:z-10"
/>
<span className="inline-flex items-center px-3 rounded-r-md border border-l-0 border-input bg-muted text-sm text-muted-foreground font-mono">
{currentFormat?.extension || ".drawio"}
</span>
</div>
</div>
</div>
<DialogFooter>
<Button
variant="outline"
onClick={() => onOpenChange(false)}
>
Cancel
</Button>
<Button onClick={handleSave}>Save</Button>
</DialogFooter>
</DialogContent>
</Dialog>
)
}

View File

@@ -0,0 +1,436 @@
"use client"
import { Moon, Sun } from "lucide-react"
import { useEffect, useState } from "react"
import { Button } from "@/components/ui/button"
import {
Dialog,
DialogContent,
DialogDescription,
DialogHeader,
DialogTitle,
} from "@/components/ui/dialog"
import { Input } from "@/components/ui/input"
import { Label } from "@/components/ui/label"
import {
Select,
SelectContent,
SelectItem,
SelectTrigger,
SelectValue,
} from "@/components/ui/select"
import { Switch } from "@/components/ui/switch"
interface SettingsDialogProps {
open: boolean
onOpenChange: (open: boolean) => void
onCloseProtectionChange?: (enabled: boolean) => void
drawioUi: "min" | "sketch"
onToggleDrawioUi: () => void
darkMode: boolean
onToggleDarkMode: () => void
}
export const STORAGE_ACCESS_CODE_KEY = "next-ai-draw-io-access-code"
export const STORAGE_CLOSE_PROTECTION_KEY = "next-ai-draw-io-close-protection"
const STORAGE_ACCESS_CODE_REQUIRED_KEY = "next-ai-draw-io-access-code-required"
export const STORAGE_AI_PROVIDER_KEY = "next-ai-draw-io-ai-provider"
export const STORAGE_AI_BASE_URL_KEY = "next-ai-draw-io-ai-base-url"
export const STORAGE_AI_API_KEY_KEY = "next-ai-draw-io-ai-api-key"
export const STORAGE_AI_MODEL_KEY = "next-ai-draw-io-ai-model"
function getStoredAccessCodeRequired(): boolean | null {
if (typeof window === "undefined") return null
const stored = localStorage.getItem(STORAGE_ACCESS_CODE_REQUIRED_KEY)
if (stored === null) return null
return stored === "true"
}
export function SettingsDialog({
open,
onOpenChange,
onCloseProtectionChange,
drawioUi,
onToggleDrawioUi,
darkMode,
onToggleDarkMode,
}: SettingsDialogProps) {
const [accessCode, setAccessCode] = useState("")
const [closeProtection, setCloseProtection] = useState(true)
const [isVerifying, setIsVerifying] = useState(false)
const [error, setError] = useState("")
const [accessCodeRequired, setAccessCodeRequired] = useState(
() => getStoredAccessCodeRequired() ?? false,
)
const [provider, setProvider] = useState("")
const [baseUrl, setBaseUrl] = useState("")
const [apiKey, setApiKey] = useState("")
const [modelId, setModelId] = useState("")
useEffect(() => {
// Only fetch if not cached in localStorage
if (getStoredAccessCodeRequired() !== null) return
fetch("/api/config")
.then((res) => {
if (!res.ok) throw new Error(`HTTP ${res.status}`)
return res.json()
})
.then((data) => {
const required = data?.accessCodeRequired === true
localStorage.setItem(
STORAGE_ACCESS_CODE_REQUIRED_KEY,
String(required),
)
setAccessCodeRequired(required)
})
.catch(() => {
// Don't cache on error - allow retry on next mount
setAccessCodeRequired(false)
})
}, [])
useEffect(() => {
if (open) {
const storedCode =
localStorage.getItem(STORAGE_ACCESS_CODE_KEY) || ""
setAccessCode(storedCode)
const storedCloseProtection = localStorage.getItem(
STORAGE_CLOSE_PROTECTION_KEY,
)
// Default to true if not set
setCloseProtection(storedCloseProtection !== "false")
// Load AI provider settings
setProvider(localStorage.getItem(STORAGE_AI_PROVIDER_KEY) || "")
setBaseUrl(localStorage.getItem(STORAGE_AI_BASE_URL_KEY) || "")
setApiKey(localStorage.getItem(STORAGE_AI_API_KEY_KEY) || "")
setModelId(localStorage.getItem(STORAGE_AI_MODEL_KEY) || "")
setError("")
}
}, [open])
const handleSave = async () => {
if (!accessCodeRequired) return
setError("")
setIsVerifying(true)
try {
const response = await fetch("/api/verify-access-code", {
method: "POST",
headers: {
"x-access-code": accessCode.trim(),
},
})
const data = await response.json()
if (!data.valid) {
setError(data.message || "Invalid access code")
return
}
localStorage.setItem(STORAGE_ACCESS_CODE_KEY, accessCode.trim())
onOpenChange(false)
} catch {
setError("Failed to verify access code")
} finally {
setIsVerifying(false)
}
}
const handleKeyDown = (e: React.KeyboardEvent) => {
if (e.key === "Enter") {
e.preventDefault()
handleSave()
}
}
return (
<Dialog open={open} onOpenChange={onOpenChange}>
<DialogContent className="sm:max-w-md">
<DialogHeader>
<DialogTitle>Settings</DialogTitle>
<DialogDescription>
Configure your application settings.
</DialogDescription>
</DialogHeader>
<div className="space-y-4 py-2">
{accessCodeRequired && (
<div className="space-y-2">
<Label htmlFor="access-code">Access Code</Label>
<div className="flex gap-2">
<Input
id="access-code"
type="password"
value={accessCode}
onChange={(e) =>
setAccessCode(e.target.value)
}
onKeyDown={handleKeyDown}
placeholder="Enter access code"
autoComplete="off"
/>
<Button
onClick={handleSave}
disabled={isVerifying || !accessCode.trim()}
>
{isVerifying ? "..." : "Save"}
</Button>
</div>
<p className="text-[0.8rem] text-muted-foreground">
Required to use this application.
</p>
{error && (
<p className="text-[0.8rem] text-destructive">
{error}
</p>
)}
</div>
)}
<div className="space-y-2">
<Label>AI Provider Settings</Label>
<p className="text-[0.8rem] text-muted-foreground">
Use your own API key to bypass usage limits. Your
key is stored locally in your browser and is never
stored on the server.
</p>
<div className="space-y-3 pt-2">
<div className="space-y-2">
<Label htmlFor="ai-provider">Provider</Label>
<Select
value={provider || "default"}
onValueChange={(value) => {
const actualValue =
value === "default" ? "" : value
setProvider(actualValue)
localStorage.setItem(
STORAGE_AI_PROVIDER_KEY,
actualValue,
)
}}
>
<SelectTrigger id="ai-provider">
<SelectValue placeholder="Use Server Default" />
</SelectTrigger>
<SelectContent>
<SelectItem value="default">
Use Server Default
</SelectItem>
<SelectItem value="openai">
OpenAI
</SelectItem>
<SelectItem value="anthropic">
Anthropic
</SelectItem>
<SelectItem value="google">
Google
</SelectItem>
<SelectItem value="azure">
Azure OpenAI
</SelectItem>
<SelectItem value="openrouter">
OpenRouter
</SelectItem>
<SelectItem value="deepseek">
DeepSeek
</SelectItem>
<SelectItem value="siliconflow">
SiliconFlow
</SelectItem>
</SelectContent>
</Select>
</div>
{provider && provider !== "default" && (
<>
<div className="space-y-2">
<Label htmlFor="ai-model">
Model ID
</Label>
<Input
id="ai-model"
value={modelId}
onChange={(e) => {
setModelId(e.target.value)
localStorage.setItem(
STORAGE_AI_MODEL_KEY,
e.target.value,
)
}}
placeholder={
provider === "openai"
? "e.g., gpt-4o"
: provider === "anthropic"
? "e.g., claude-sonnet-4-5"
: provider === "google"
? "e.g., gemini-2.0-flash-exp"
: provider ===
"deepseek"
? "e.g., deepseek-chat"
: "Model ID"
}
/>
</div>
<div className="space-y-2">
<Label htmlFor="ai-api-key">
API Key
</Label>
<Input
id="ai-api-key"
type="password"
value={apiKey}
onChange={(e) => {
setApiKey(e.target.value)
localStorage.setItem(
STORAGE_AI_API_KEY_KEY,
e.target.value,
)
}}
placeholder="Your API key"
autoComplete="off"
/>
<p className="text-[0.8rem] text-muted-foreground">
Overrides{" "}
{provider === "openai"
? "OPENAI_API_KEY"
: provider === "anthropic"
? "ANTHROPIC_API_KEY"
: provider === "google"
? "GOOGLE_GENERATIVE_AI_API_KEY"
: provider === "azure"
? "AZURE_API_KEY"
: provider ===
"openrouter"
? "OPENROUTER_API_KEY"
: provider ===
"deepseek"
? "DEEPSEEK_API_KEY"
: provider ===
"siliconflow"
? "SILICONFLOW_API_KEY"
: "server API key"}
</p>
</div>
<div className="space-y-2">
<Label htmlFor="ai-base-url">
Base URL (optional)
</Label>
<Input
id="ai-base-url"
value={baseUrl}
onChange={(e) => {
setBaseUrl(e.target.value)
localStorage.setItem(
STORAGE_AI_BASE_URL_KEY,
e.target.value,
)
}}
placeholder={
provider === "anthropic"
? "https://api.anthropic.com/v1"
: provider === "siliconflow"
? "https://api.siliconflow.com/v1"
: "Custom endpoint URL"
}
/>
</div>
<Button
variant="outline"
size="sm"
className="w-full"
onClick={() => {
localStorage.removeItem(
STORAGE_AI_PROVIDER_KEY,
)
localStorage.removeItem(
STORAGE_AI_BASE_URL_KEY,
)
localStorage.removeItem(
STORAGE_AI_API_KEY_KEY,
)
localStorage.removeItem(
STORAGE_AI_MODEL_KEY,
)
setProvider("")
setBaseUrl("")
setApiKey("")
setModelId("")
}}
>
Clear Settings
</Button>
</>
)}
</div>
</div>
<div className="flex items-center justify-between">
<div className="space-y-0.5">
<Label htmlFor="theme-toggle">Theme</Label>
<p className="text-[0.8rem] text-muted-foreground">
Dark/Light mode for interface and DrawIO canvas.
</p>
</div>
<Button
id="theme-toggle"
variant="outline"
size="icon"
onClick={onToggleDarkMode}
>
{darkMode ? (
<Sun className="h-4 w-4" />
) : (
<Moon className="h-4 w-4" />
)}
</Button>
</div>
<div className="flex items-center justify-between">
<div className="space-y-0.5">
<Label htmlFor="drawio-ui">DrawIO Style</Label>
<p className="text-[0.8rem] text-muted-foreground">
Canvas style:{" "}
{drawioUi === "min" ? "Minimal" : "Sketch"}
</p>
</div>
<Button
id="drawio-ui"
variant="outline"
size="sm"
onClick={onToggleDrawioUi}
>
Switch to{" "}
{drawioUi === "min" ? "Sketch" : "Minimal"}
</Button>
</div>
<div className="flex items-center justify-between">
<div className="space-y-0.5">
<Label htmlFor="close-protection">
Close Protection
</Label>
<p className="text-[0.8rem] text-muted-foreground">
Show confirmation when leaving the page.
</p>
</div>
<Switch
id="close-protection"
checked={closeProtection}
onCheckedChange={(checked) => {
setCloseProtection(checked)
localStorage.setItem(
STORAGE_CLOSE_PROTECTION_KEY,
checked.toString(),
)
onCloseProtectionChange?.(checked)
}}
/>
</div>
</div>
</DialogContent>
</Dialog>
)
}

View File

@@ -10,7 +10,7 @@ const buttonVariants = cva(
variants: {
variant: {
default:
"bg-primary text-primary-foreground shadow-xs hover:bg-primary/90",
"bg-primary text-primary-foreground shadow-xs hover:brightness-75",
destructive:
"bg-destructive text-white shadow-xs hover:bg-destructive/90 focus-visible:ring-destructive/20 dark:focus-visible:ring-destructive/40 dark:bg-destructive/60",
outline:

View File

@@ -0,0 +1,33 @@
"use client"
import * as CollapsiblePrimitive from "@radix-ui/react-collapsible"
function Collapsible({
...props
}: React.ComponentProps<typeof CollapsiblePrimitive.Root>) {
return <CollapsiblePrimitive.Root data-slot="collapsible" {...props} />
}
function CollapsibleTrigger({
...props
}: React.ComponentProps<typeof CollapsiblePrimitive.CollapsibleTrigger>) {
return (
<CollapsiblePrimitive.CollapsibleTrigger
data-slot="collapsible-trigger"
{...props}
/>
)
}
function CollapsibleContent({
...props
}: React.ComponentProps<typeof CollapsiblePrimitive.CollapsibleContent>) {
return (
<CollapsiblePrimitive.CollapsibleContent
data-slot="collapsible-content"
{...props}
/>
)
}
export { Collapsible, CollapsibleTrigger, CollapsibleContent }

24
components/ui/label.tsx Normal file
View File

@@ -0,0 +1,24 @@
"use client"
import * as React from "react"
import * as LabelPrimitive from "@radix-ui/react-label"
import { cn } from "@/lib/utils"
function Label({
className,
...props
}: React.ComponentProps<typeof LabelPrimitive.Root>) {
return (
<LabelPrimitive.Root
data-slot="label"
className={cn(
"flex items-center gap-2 text-sm leading-none font-medium select-none group-data-[disabled=true]:pointer-events-none group-data-[disabled=true]:opacity-50 peer-disabled:cursor-not-allowed peer-disabled:opacity-50",
className
)}
{...props}
/>
)
}
export { Label }

View File

@@ -0,0 +1,56 @@
"use client"
import * as React from "react"
import { GripVerticalIcon } from "lucide-react"
import * as ResizablePrimitive from "react-resizable-panels"
import { cn } from "@/lib/utils"
function ResizablePanelGroup({
className,
...props
}: React.ComponentProps<typeof ResizablePrimitive.PanelGroup>) {
return (
<ResizablePrimitive.PanelGroup
data-slot="resizable-panel-group"
className={cn(
"flex h-full w-full data-[panel-group-direction=vertical]:flex-col",
className
)}
{...props}
/>
)
}
function ResizablePanel({
...props
}: React.ComponentProps<typeof ResizablePrimitive.Panel>) {
return <ResizablePrimitive.Panel data-slot="resizable-panel" {...props} />
}
function ResizableHandle({
withHandle,
className,
...props
}: React.ComponentProps<typeof ResizablePrimitive.PanelResizeHandle> & {
withHandle?: boolean
}) {
return (
<ResizablePrimitive.PanelResizeHandle
data-slot="resizable-handle"
className={cn(
"bg-border focus-visible:ring-ring relative flex w-px items-center justify-center after:absolute after:inset-y-0 after:left-1/2 after:w-1 after:-translate-x-1/2 focus-visible:ring-1 focus-visible:ring-offset-1 focus-visible:outline-hidden data-[panel-group-direction=vertical]:h-px data-[panel-group-direction=vertical]:w-full data-[panel-group-direction=vertical]:after:left-0 data-[panel-group-direction=vertical]:after:h-1 data-[panel-group-direction=vertical]:after:w-full data-[panel-group-direction=vertical]:after:translate-x-0 data-[panel-group-direction=vertical]:after:-translate-y-1/2 [&[data-panel-group-direction=vertical]>div]:rotate-90",
className
)}
{...props}
>
{withHandle && (
<div className="bg-border z-10 flex h-4 w-3 items-center justify-center rounded-xs border">
<GripVerticalIcon className="size-2.5" />
</div>
)}
</ResizablePrimitive.PanelResizeHandle>
)
}
export { ResizablePanelGroup, ResizablePanel, ResizableHandle }

View File

@@ -18,7 +18,7 @@ function ScrollArea({
>
<ScrollAreaPrimitive.Viewport
data-slot="scroll-area-viewport"
className="ring-ring/10 dark:ring-ring/20 dark:outline-ring/40 outline-ring/50 size-full rounded-[inherit] transition-[color,box-shadow] focus-visible:ring-4 focus-visible:outline-1"
className="ring-ring/10 dark:ring-ring/20 dark:outline-ring/40 outline-ring/50 size-full rounded-[inherit] transition-[color,box-shadow] focus-visible:ring-4 focus-visible:outline-1 !overflow-x-hidden"
>
{children}
</ScrollAreaPrimitive.Viewport>

187
components/ui/select.tsx Normal file
View File

@@ -0,0 +1,187 @@
"use client"
import * as React from "react"
import * as SelectPrimitive from "@radix-ui/react-select"
import { CheckIcon, ChevronDownIcon, ChevronUpIcon } from "lucide-react"
import { cn } from "@/lib/utils"
function Select({
...props
}: React.ComponentProps<typeof SelectPrimitive.Root>) {
return <SelectPrimitive.Root data-slot="select" {...props} />
}
function SelectGroup({
...props
}: React.ComponentProps<typeof SelectPrimitive.Group>) {
return <SelectPrimitive.Group data-slot="select-group" {...props} />
}
function SelectValue({
...props
}: React.ComponentProps<typeof SelectPrimitive.Value>) {
return <SelectPrimitive.Value data-slot="select-value" {...props} />
}
function SelectTrigger({
className,
size = "default",
children,
...props
}: React.ComponentProps<typeof SelectPrimitive.Trigger> & {
size?: "sm" | "default"
}) {
return (
<SelectPrimitive.Trigger
data-slot="select-trigger"
data-size={size}
className={cn(
"border-input data-[placeholder]:text-muted-foreground [&_svg:not([class*='text-'])]:text-muted-foreground focus-visible:border-ring focus-visible:ring-ring/50 aria-invalid:ring-destructive/20 dark:aria-invalid:ring-destructive/40 aria-invalid:border-destructive dark:bg-input/30 dark:hover:bg-input/50 flex w-fit items-center justify-between gap-2 rounded-md border bg-transparent px-3 py-2 text-sm whitespace-nowrap shadow-xs transition-[color,box-shadow] outline-none focus-visible:ring-[3px] disabled:cursor-not-allowed disabled:opacity-50 data-[size=default]:h-9 data-[size=sm]:h-8 *:data-[slot=select-value]:line-clamp-1 *:data-[slot=select-value]:flex *:data-[slot=select-value]:items-center *:data-[slot=select-value]:gap-2 [&_svg]:pointer-events-none [&_svg]:shrink-0 [&_svg:not([class*='size-'])]:size-4",
className
)}
{...props}
>
{children}
<SelectPrimitive.Icon asChild>
<ChevronDownIcon className="size-4 opacity-50" />
</SelectPrimitive.Icon>
</SelectPrimitive.Trigger>
)
}
function SelectContent({
className,
children,
position = "popper",
align = "center",
...props
}: React.ComponentProps<typeof SelectPrimitive.Content>) {
return (
<SelectPrimitive.Portal>
<SelectPrimitive.Content
data-slot="select-content"
className={cn(
"bg-popover text-popover-foreground data-[state=open]:animate-in data-[state=closed]:animate-out data-[state=closed]:fade-out-0 data-[state=open]:fade-in-0 data-[state=closed]:zoom-out-95 data-[state=open]:zoom-in-95 data-[side=bottom]:slide-in-from-top-2 data-[side=left]:slide-in-from-right-2 data-[side=right]:slide-in-from-left-2 data-[side=top]:slide-in-from-bottom-2 relative z-50 max-h-(--radix-select-content-available-height) min-w-[8rem] origin-(--radix-select-content-transform-origin) overflow-x-hidden overflow-y-auto rounded-md border shadow-md",
position === "popper" &&
"data-[side=bottom]:translate-y-1 data-[side=left]:-translate-x-1 data-[side=right]:translate-x-1 data-[side=top]:-translate-y-1",
className
)}
position={position}
align={align}
{...props}
>
<SelectScrollUpButton />
<SelectPrimitive.Viewport
className={cn(
"p-1",
position === "popper" &&
"h-[var(--radix-select-trigger-height)] w-full min-w-[var(--radix-select-trigger-width)] scroll-my-1"
)}
>
{children}
</SelectPrimitive.Viewport>
<SelectScrollDownButton />
</SelectPrimitive.Content>
</SelectPrimitive.Portal>
)
}
function SelectLabel({
className,
...props
}: React.ComponentProps<typeof SelectPrimitive.Label>) {
return (
<SelectPrimitive.Label
data-slot="select-label"
className={cn("text-muted-foreground px-2 py-1.5 text-xs", className)}
{...props}
/>
)
}
function SelectItem({
className,
children,
...props
}: React.ComponentProps<typeof SelectPrimitive.Item>) {
return (
<SelectPrimitive.Item
data-slot="select-item"
className={cn(
"focus:bg-accent focus:text-accent-foreground [&_svg:not([class*='text-'])]:text-muted-foreground relative flex w-full cursor-default items-center gap-2 rounded-sm py-1.5 pr-8 pl-2 text-sm outline-hidden select-none data-[disabled]:pointer-events-none data-[disabled]:opacity-50 [&_svg]:pointer-events-none [&_svg]:shrink-0 [&_svg:not([class*='size-'])]:size-4 *:[span]:last:flex *:[span]:last:items-center *:[span]:last:gap-2",
className
)}
{...props}
>
<span className="absolute right-2 flex size-3.5 items-center justify-center">
<SelectPrimitive.ItemIndicator>
<CheckIcon className="size-4" />
</SelectPrimitive.ItemIndicator>
</span>
<SelectPrimitive.ItemText>{children}</SelectPrimitive.ItemText>
</SelectPrimitive.Item>
)
}
function SelectSeparator({
className,
...props
}: React.ComponentProps<typeof SelectPrimitive.Separator>) {
return (
<SelectPrimitive.Separator
data-slot="select-separator"
className={cn("bg-border pointer-events-none -mx-1 my-1 h-px", className)}
{...props}
/>
)
}
function SelectScrollUpButton({
className,
...props
}: React.ComponentProps<typeof SelectPrimitive.ScrollUpButton>) {
return (
<SelectPrimitive.ScrollUpButton
data-slot="select-scroll-up-button"
className={cn(
"flex cursor-default items-center justify-center py-1",
className
)}
{...props}
>
<ChevronUpIcon className="size-4" />
</SelectPrimitive.ScrollUpButton>
)
}
function SelectScrollDownButton({
className,
...props
}: React.ComponentProps<typeof SelectPrimitive.ScrollDownButton>) {
return (
<SelectPrimitive.ScrollDownButton
data-slot="select-scroll-down-button"
className={cn(
"flex cursor-default items-center justify-center py-1",
className
)}
{...props}
>
<ChevronDownIcon className="size-4" />
</SelectPrimitive.ScrollDownButton>
)
}
export {
Select,
SelectContent,
SelectGroup,
SelectItem,
SelectLabel,
SelectScrollDownButton,
SelectScrollUpButton,
SelectSeparator,
SelectTrigger,
SelectValue,
}

31
components/ui/switch.tsx Normal file
View File

@@ -0,0 +1,31 @@
"use client"
import * as React from "react"
import * as SwitchPrimitive from "@radix-ui/react-switch"
import { cn } from "@/lib/utils"
function Switch({
className,
...props
}: React.ComponentProps<typeof SwitchPrimitive.Root>) {
return (
<SwitchPrimitive.Root
data-slot="switch"
className={cn(
"peer data-[state=checked]:bg-primary data-[state=unchecked]:bg-input focus-visible:border-ring focus-visible:ring-ring/50 dark:data-[state=unchecked]:bg-input/80 inline-flex h-[1.15rem] w-8 shrink-0 items-center rounded-full border border-transparent shadow-xs transition-all outline-none focus-visible:ring-[3px] disabled:cursor-not-allowed disabled:opacity-50",
className
)}
{...props}
>
<SwitchPrimitive.Thumb
data-slot="switch-thumb"
className={cn(
"bg-background dark:data-[state=unchecked]:bg-foreground dark:data-[state=checked]:bg-primary-foreground pointer-events-none block size-4 rounded-full ring-0 transition-transform data-[state=checked]:translate-x-[calc(100%-2px)] data-[state=unchecked]:translate-x-0"
)}
/>
</SwitchPrimitive.Root>
)
}
export { Switch }

View File

@@ -1,72 +1,277 @@
"use client";
"use client"
import React, { createContext, useContext, useRef, useState } from "react";
import type { DrawIoEmbedRef } from "react-drawio";
import { extractDiagramXML } from "../lib/utils";
import type React from "react"
import { createContext, useContext, useRef, useState } from "react"
import type { DrawIoEmbedRef } from "react-drawio"
import { STORAGE_DIAGRAM_XML_KEY } from "@/components/chat-panel"
import type { ExportFormat } from "@/components/save-dialog"
import { extractDiagramXML, validateAndFixXml } from "../lib/utils"
interface DiagramContextType {
chartXML: string;
latestSvg: string;
diagramHistory: { svg: string; xml: string }[];
loadDiagram: (chart: string) => void;
handleExport: () => void;
resolverRef: React.Ref<((value: string) => void) | null>;
drawioRef: React.Ref<DrawIoEmbedRef | null>;
handleDiagramExport: (data: any) => void;
clearDiagram: () => void;
chartXML: string
latestSvg: string
diagramHistory: { svg: string; xml: string }[]
loadDiagram: (chart: string, skipValidation?: boolean) => string | null
handleExport: () => void
handleExportWithoutHistory: () => void
resolverRef: React.Ref<((value: string) => void) | null>
drawioRef: React.Ref<DrawIoEmbedRef | null>
handleDiagramExport: (data: any) => void
clearDiagram: () => void
saveDiagramToFile: (
filename: string,
format: ExportFormat,
sessionId?: string,
) => void
isDrawioReady: boolean
onDrawioLoad: () => void
resetDrawioReady: () => void
}
const DiagramContext = createContext<DiagramContextType | undefined>(undefined);
const DiagramContext = createContext<DiagramContextType | undefined>(undefined)
export function DiagramProvider({ children }: { children: React.ReactNode }) {
const [chartXML, setChartXML] = useState<string>("");
const [latestSvg, setLatestSvg] = useState<string>("");
const [chartXML, setChartXML] = useState<string>("")
const [latestSvg, setLatestSvg] = useState<string>("")
const [diagramHistory, setDiagramHistory] = useState<
{ svg: string; xml: string }[]
>([]);
const drawioRef = useRef<DrawIoEmbedRef | null>(null);
const resolverRef = useRef<((value: string) => void) | null>(null);
>([])
const [isDrawioReady, setIsDrawioReady] = useState(false)
const hasCalledOnLoadRef = useRef(false)
const drawioRef = useRef<DrawIoEmbedRef | null>(null)
const resolverRef = useRef<((value: string) => void) | null>(null)
// Track if we're expecting an export for history (user-initiated)
const expectHistoryExportRef = useRef<boolean>(false)
const onDrawioLoad = () => {
// Only set ready state once to prevent infinite loops
if (hasCalledOnLoadRef.current) return
hasCalledOnLoadRef.current = true
// console.log("[DiagramContext] DrawIO loaded, setting ready state")
setIsDrawioReady(true)
}
const resetDrawioReady = () => {
// console.log("[DiagramContext] Resetting DrawIO ready state")
hasCalledOnLoadRef.current = false
setIsDrawioReady(false)
}
// Track if we're expecting an export for file save (stores raw export data)
const saveResolverRef = useRef<{
resolver: ((data: string) => void) | null
format: ExportFormat | null
}>({ resolver: null, format: null })
const handleExport = () => {
if (drawioRef.current) {
// Mark that this export should be saved to history
expectHistoryExportRef.current = true
drawioRef.current.exportDiagram({
format: "xmlsvg",
});
})
}
};
}
const loadDiagram = (chart: string) => {
const handleExportWithoutHistory = () => {
if (drawioRef.current) {
drawioRef.current.load({
xml: chart,
});
// Export without saving to history (for edit_diagram fetching current state)
drawioRef.current.exportDiagram({
format: "xmlsvg",
})
}
};
}
const loadDiagram = (
chart: string,
skipValidation?: boolean,
): string | null => {
console.time("perf:loadDiagram")
let xmlToLoad = chart
// Validate XML structure before loading (unless skipped for internal use)
if (!skipValidation) {
console.time("perf:loadDiagram-validation")
const validation = validateAndFixXml(chart)
console.timeEnd("perf:loadDiagram-validation")
if (!validation.valid) {
console.warn(
"[loadDiagram] Validation error:",
validation.error,
)
console.timeEnd("perf:loadDiagram")
return validation.error
}
// Use fixed XML if auto-fix was applied
if (validation.fixed) {
console.log(
"[loadDiagram] Auto-fixed XML issues:",
validation.fixes,
)
xmlToLoad = validation.fixed
}
}
// Keep chartXML in sync even when diagrams are injected (e.g., display_diagram tool)
setChartXML(xmlToLoad)
if (drawioRef.current) {
console.time("perf:drawio-iframe-load")
drawioRef.current.load({
xml: xmlToLoad,
})
console.timeEnd("perf:drawio-iframe-load")
}
console.timeEnd("perf:loadDiagram")
return null
}
const handleDiagramExport = (data: any) => {
const extractedXML = extractDiagramXML(data.data);
setChartXML(extractedXML);
setLatestSvg(data.data);
setDiagramHistory((prev) => [
...prev,
{
svg: data.data,
xml: extractedXML,
},
]);
if (resolverRef.current) {
resolverRef.current(extractedXML);
resolverRef.current = null;
// Handle save to file if requested (process raw data before extraction)
if (saveResolverRef.current.resolver) {
const format = saveResolverRef.current.format
saveResolverRef.current.resolver(data.data)
saveResolverRef.current = { resolver: null, format: null }
// For non-xmlsvg formats, skip XML extraction as it will fail
// Only drawio (which uses xmlsvg internally) has the content attribute
if (format === "png" || format === "svg") {
return
}
}
};
const extractedXML = extractDiagramXML(data.data)
setChartXML(extractedXML)
setLatestSvg(data.data)
// Only add to history if this was a user-initiated export
// Limit to 20 entries to prevent memory leaks during long sessions
const MAX_HISTORY_SIZE = 20
if (expectHistoryExportRef.current) {
setDiagramHistory((prev) => {
const newHistory = [
...prev,
{
svg: data.data,
xml: extractedXML,
},
]
// Keep only the last MAX_HISTORY_SIZE entries (circular buffer)
return newHistory.slice(-MAX_HISTORY_SIZE)
})
expectHistoryExportRef.current = false
}
if (resolverRef.current) {
resolverRef.current(extractedXML)
resolverRef.current = null
}
}
const clearDiagram = () => {
const emptyDiagram = `<mxfile><diagram name="Page-1" id="page-1"><mxGraphModel><root><mxCell id="0"/><mxCell id="1" parent="0"/></root></mxGraphModel></diagram></mxfile>`;
loadDiagram(emptyDiagram);
setChartXML(emptyDiagram);
setLatestSvg("");
setDiagramHistory([]);
};
const emptyDiagram = `<mxfile><diagram name="Page-1" id="page-1"><mxGraphModel><root><mxCell id="0"/><mxCell id="1" parent="0"/></root></mxGraphModel></diagram></mxfile>`
// Skip validation for trusted internal template (loadDiagram also sets chartXML)
loadDiagram(emptyDiagram, true)
setLatestSvg("")
setDiagramHistory([])
}
const saveDiagramToFile = (
filename: string,
format: ExportFormat,
sessionId?: string,
) => {
if (!drawioRef.current) {
console.warn("Draw.io editor not ready")
return
}
// Map format to draw.io export format
const drawioFormat = format === "drawio" ? "xmlsvg" : format
// Set up the resolver before triggering export
saveResolverRef.current = {
resolver: (exportData: string) => {
let fileContent: string | Blob
let mimeType: string
let extension: string
if (format === "drawio") {
// Extract XML from SVG for .drawio format
const xml = extractDiagramXML(exportData)
let xmlContent = xml
if (!xml.includes("<mxfile")) {
xmlContent = `<mxfile><diagram name="Page-1" id="page-1">${xml}</diagram></mxfile>`
}
fileContent = xmlContent
mimeType = "application/xml"
extension = ".drawio"
// Save to localStorage when user manually saves
localStorage.setItem(STORAGE_DIAGRAM_XML_KEY, xmlContent)
} else if (format === "png") {
// PNG data comes as base64 data URL
fileContent = exportData
mimeType = "image/png"
extension = ".png"
} else {
// SVG format
fileContent = exportData
mimeType = "image/svg+xml"
extension = ".svg"
}
// Log save event to Langfuse (flags the trace)
logSaveToLangfuse(filename, format, sessionId)
// Handle download
let url: string
if (
typeof fileContent === "string" &&
fileContent.startsWith("data:")
) {
// Already a data URL (PNG)
url = fileContent
} else {
const blob = new Blob([fileContent], { type: mimeType })
url = URL.createObjectURL(blob)
}
const a = document.createElement("a")
a.href = url
a.download = `${filename}${extension}`
document.body.appendChild(a)
a.click()
document.body.removeChild(a)
// Delay URL revocation to ensure download completes
if (!url.startsWith("data:")) {
setTimeout(() => URL.revokeObjectURL(url), 100)
}
},
format,
}
// Export diagram - callback will be handled in handleDiagramExport
drawioRef.current.exportDiagram({ format: drawioFormat })
}
// Log save event to Langfuse (just flags the trace, doesn't send content)
const logSaveToLangfuse = async (
filename: string,
format: string,
sessionId?: string,
) => {
try {
await fetch("/api/log-save", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ filename, format, sessionId }),
})
} catch (error) {
console.warn("Failed to log save to Langfuse:", error)
}
}
return (
<DiagramContext.Provider
@@ -76,21 +281,26 @@ export function DiagramProvider({ children }: { children: React.ReactNode }) {
diagramHistory,
loadDiagram,
handleExport,
handleExportWithoutHistory,
resolverRef,
drawioRef,
handleDiagramExport,
clearDiagram,
saveDiagramToFile,
isDrawioReady,
onDrawioLoad,
resetDrawioReady,
}}
>
{children}
</DiagramContext.Provider>
);
)
}
export function useDiagram() {
const context = useContext(DiagramContext);
const context = useContext(DiagramContext)
if (context === undefined) {
throw new Error("useDiagram must be used within a DiagramProvider");
throw new Error("useDiagram must be used within a DiagramProvider")
}
return context;
return context
}

12
docker-compose.yml Normal file
View File

@@ -0,0 +1,12 @@
services:
drawio:
image: jgraph/drawio:latest
ports: ["8080:8080"]
next-ai-draw-io:
build:
context: .
args:
- NEXT_PUBLIC_DRAWIO_BASE_URL=http://localhost:8080
ports: ["3000:3000"]
env_file: .env
depends_on: [drawio]

250
docs/README_CN.md Normal file
View File

@@ -0,0 +1,250 @@
# Next AI Draw.io
<div align="center">
**AI驱动的图表创建工具 - 对话、绘制、可视化**
[English](../README.md) | 中文 | [日本語](./README_JA.md)
[![TrendShift](https://trendshift.io/api/badge/repositories/15449)](https://next-ai-drawio.jiang.jp/)
[![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
[![Next.js](https://img.shields.io/badge/Next.js-16.x-black)](https://nextjs.org/)
[![React](https://img.shields.io/badge/React-19.x-61dafb)](https://react.dev/)
[![Sponsor](https://img.shields.io/badge/Sponsor-❤-ea4aaa)](https://github.com/sponsors/DayuanJiang)
[![Live Demo](../public/live-demo-button.svg)](https://next-ai-drawio.jiang.jp/)
</div>
一个集成了AI功能的Next.js网页应用与draw.io图表无缝结合。通过自然语言命令和AI辅助可视化来创建、修改和增强图表。
https://github.com/user-attachments/assets/b2eef5f3-b335-4e71-a755-dc2e80931979
## 目录
- [Next AI Draw.io](#next-ai-drawio)
- [目录](#目录)
- [示例](#示例)
- [功能特性](#功能特性)
- [快速开始](#快速开始)
- [在线试用](#在线试用)
- [使用Docker运行推荐](#使用docker运行推荐)
- [安装](#安装)
- [部署](#部署)
- [多提供商支持](#多提供商支持)
- [工作原理](#工作原理)
- [项目结构](#项目结构)
- [支持与联系](#支持与联系)
- [Star历史](#star历史)
## 示例
以下是一些示例提示词及其生成的图表:
<div align="center">
<table width="100%">
<tr>
<td colspan="2" valign="top" align="center">
<strong>动画Transformer连接器</strong><br />
<p><strong>提示词:</strong> 给我一个带有**动画连接器**的Transformer架构图。</p>
<img src="../public/animated_connectors.svg" alt="带动画连接器的Transformer架构" width="480" />
</td>
</tr>
<tr>
<td width="50%" valign="top">
<strong>GCP架构图</strong><br />
<p><strong>提示词:</strong> 使用**GCP图标**生成一个GCP架构图。在这个图中用户连接到托管在实例上的前端。</p>
<img src="../public/gcp_demo.svg" alt="GCP架构图" width="480" />
</td>
<td width="50%" valign="top">
<strong>AWS架构图</strong><br />
<p><strong>提示词:</strong> 使用**AWS图标**生成一个AWS架构图。在这个图中用户连接到托管在实例上的前端。</p>
<img src="../public/aws_demo.svg" alt="AWS架构图" width="480" />
</td>
</tr>
<tr>
<td width="50%" valign="top">
<strong>Azure架构图</strong><br />
<p><strong>提示词:</strong> 使用**Azure图标**生成一个Azure架构图。在这个图中用户连接到托管在实例上的前端。</p>
<img src="../public/azure_demo.svg" alt="Azure架构图" width="480" />
</td>
<td width="50%" valign="top">
<strong>猫咪素描</strong><br />
<p><strong>提示词:</strong> 给我画一只可爱的猫。</p>
<img src="../public/cat_demo.svg" alt="猫咪绘图" width="240" />
</td>
</tr>
</table>
</div>
## 功能特性
- **LLM驱动的图表创建**利用大语言模型通过自然语言命令直接创建和操作draw.io图表
- **基于图像的图表复制**上传现有图表或图像让AI自动复制和增强
- **PDF和文本文件上传**上传PDF文档和文本文件提取内容并从现有文档生成图表
- **AI推理过程显示**查看支持模型的AI思考过程OpenAI o1/o3、Gemini、Claude等
- **图表历史记录**全面的版本控制跟踪所有更改允许您查看和恢复AI编辑前的图表版本
- **交互式聊天界面**与AI实时对话来完善您的图表
- **云架构图支持**专门支持生成云架构图AWS、GCP、Azure
- **动画连接器**:在图表元素之间创建动态动画连接器,实现更好的可视化效果
## 快速开始
### 在线试用
无需安装!直接在我们的演示站点试用:
[![Live Demo](../public/live-demo-button.svg)](https://next-ai-drawio.jiang.jp/)
> 注意:由于访问量较大,演示站点目前使用 minimax-m2 模型。如需获得最佳效果,建议使用 Claude Sonnet 4.5 或 Claude Opus 4.5 自行部署。
> **使用自己的 API Key**:您可以使用自己的 API Key 来绕过演示站点的用量限制。点击聊天面板中的设置图标即可配置您的 Provider 和 API Key。您的 Key 仅保存在浏览器本地,不会被存储在服务器上。
### 使用Docker运行推荐
如果您只想在本地运行最好的方式是使用Docker。
首先如果您还没有安装Docker请先安装[获取Docker](https://docs.docker.com/get-docker/)
然后运行:
```bash
docker run -d -p 3000:3000 \
-e AI_PROVIDER=openai \
-e AI_MODEL=gpt-4o \
-e OPENAI_API_KEY=your_api_key \
ghcr.io/dayuanjiang/next-ai-draw-io:latest
```
或者使用 env 文件:
```bash
cp env.example .env
# 编辑 .env 填写您的配置
docker run -d -p 3000:3000 --env-file .env ghcr.io/dayuanjiang/next-ai-draw-io:latest
```
在浏览器中打开 [http://localhost:3000](http://localhost:3000)。
请根据您首选的AI提供商配置替换环境变量。可用选项请参阅[多提供商支持](#多提供商支持)。
> **离线部署:** 如果 `embed.diagrams.net` 被屏蔽,请参阅 [离线部署指南](./offline-deployment.md) 了解配置选项。
### 安装
1. 克隆仓库:
```bash
git clone https://github.com/DayuanJiang/next-ai-draw-io
cd next-ai-draw-io
```
2. 安装依赖:
```bash
npm install
```
3. 配置您的AI提供商
在根目录创建 `.env.local` 文件:
```bash
cp env.example .env.local
```
编辑 `.env.local` 并配置您选择的提供商:
-`AI_PROVIDER` 设置为您选择的提供商bedrock, openai, anthropic, google, azure, ollama, openrouter, deepseek, siliconflow
-`AI_MODEL` 设置为您要使用的特定模型
- 添加您的提供商所需的API密钥
- `TEMPERATURE`:可选的温度设置(例如 `0` 表示确定性输出)。对于不支持此参数的模型(如推理模型),请不要设置。
- `ACCESS_CODE_LIST` 访问密码,可选,可以使用逗号隔开多个密码。
> 警告:如果不填写 `ACCESS_CODE_LIST`,则任何人都可以直接使用你部署后的网站,可能会导致你的 token 被急速消耗完毕,建议填写此选项。
详细设置说明请参阅[提供商配置指南](./ai-providers.md)。
4. 运行开发服务器:
```bash
npm run dev
```
5. 在浏览器中打开 [http://localhost:3000](http://localhost:3000) 查看应用。
## 部署
部署Next.js应用最简单的方式是使用Next.js创建者提供的[Vercel平台](https://vercel.com/new)。
查看[Next.js部署文档](https://nextjs.org/docs/app/building-your-application/deploying)了解更多详情。
或者您可以通过此按钮部署:
[![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https%3A%2F%2Fgithub.com%2FDayuanJiang%2Fnext-ai-draw-io)
请确保在Vercel控制台中**设置环境变量**,就像您在本地 `.env.local` 文件中所做的那样。
## 多提供商支持
- AWS Bedrock默认
- OpenAI
- Anthropic
- Google AI
- Azure OpenAI
- Ollama
- OpenRouter
- DeepSeek
- SiliconFlow
除AWS Bedrock和OpenRouter外所有提供商都支持自定义端点。
📖 **[详细的提供商配置指南](./ai-providers.md)** - 查看各提供商的设置说明。
**模型要求**此任务需要强大的模型能力因为它涉及生成具有严格格式约束的长文本draw.io XML。推荐使用Claude Sonnet 4.5、GPT-4o、Gemini 2.0和DeepSeek V3/R1。
注意:`claude-sonnet-4-5` 已在带有AWS标志的draw.io图表上进行训练因此如果您想创建AWS架构图这是最佳选择。
## 工作原理
本应用使用以下技术:
- **Next.js**:用于前端框架和路由
- **Vercel AI SDK**`ai` + `@ai-sdk/*`用于流式AI响应和多提供商支持
- **react-drawio**:用于图表表示和操作
图表以XML格式表示可在draw.io中渲染。AI处理您的命令并相应地生成或修改此XML。
## 项目结构
```
app/ # Next.js App Router
api/chat/ # 带AI工具的聊天API端点
page.tsx # 带DrawIO嵌入的主页面
components/ # React组件
chat-panel.tsx # 带图表控制的聊天界面
chat-input.tsx # 带文件上传的用户输入组件
history-dialog.tsx # 图表版本历史查看器
ui/ # UI组件按钮、卡片等
contexts/ # React上下文提供者
diagram-context.tsx # 全局图表状态管理
lib/ # 工具函数和辅助程序
ai-providers.ts # 多提供商AI配置
utils.ts # XML处理和转换工具
public/ # 静态资源包括示例图片
```
## 支持与联系
如果您觉得这个项目有用,请考虑[赞助](https://github.com/sponsors/DayuanJiang)来帮助我托管在线演示站点!
如需支持或咨询请在GitHub仓库上提交issue或联系维护者
- 邮箱me[at]jiang.jp
## Star历史
[![Star History Chart](https://api.star-history.com/svg?repos=DayuanJiang/next-ai-draw-io&type=date&legend=top-left)](https://www.star-history.com/#DayuanJiang/next-ai-draw-io&type=date&legend=top-left)
---

250
docs/README_JA.md Normal file
View File

@@ -0,0 +1,250 @@
# Next AI Draw.io
<div align="center">
**AI搭載のダイアグラム作成ツール - チャット、描画、可視化**
[English](../README.md) | [中文](./README_CN.md) | 日本語
[![TrendShift](https://trendshift.io/api/badge/repositories/15449)](https://next-ai-drawio.jiang.jp/)
[![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
[![Next.js](https://img.shields.io/badge/Next.js-16.x-black)](https://nextjs.org/)
[![React](https://img.shields.io/badge/React-19.x-61dafb)](https://react.dev/)
[![Sponsor](https://img.shields.io/badge/Sponsor-❤-ea4aaa)](https://github.com/sponsors/DayuanJiang)
[![Live Demo](../public/live-demo-button.svg)](https://next-ai-drawio.jiang.jp/)
</div>
AI機能とdraw.ioダイアグラムを統合したNext.jsウェブアプリケーションです。自然言語コマンドとAI支援の可視化により、ダイアグラムを作成、修正、強化できます。
https://github.com/user-attachments/assets/b2eef5f3-b335-4e71-a755-dc2e80931979
## 目次
- [Next AI Draw.io](#next-ai-drawio)
- [目次](#目次)
- [](#例)
- [機能](#機能)
- [はじめに](#はじめに)
- [オンラインで試す](#オンラインで試す)
- [Dockerで実行推奨](#dockerで実行推奨)
- [インストール](#インストール)
- [デプロイ](#デプロイ)
- [マルチプロバイダーサポート](#マルチプロバイダーサポート)
- [仕組み](#仕組み)
- [プロジェクト構造](#プロジェクト構造)
- [サポート&お問い合わせ](#サポートお問い合わせ)
- [スター履歴](#スター履歴)
## 例
以下はいくつかのプロンプト例と生成されたダイアグラムです:
<div align="center">
<table width="100%">
<tr>
<td colspan="2" valign="top" align="center">
<strong>アニメーションTransformerコネクタ</strong><br />
<p><strong>プロンプト:</strong> **アニメーションコネクタ**付きのTransformerアーキテクチャ図を作成してください。</p>
<img src="../public/animated_connectors.svg" alt="アニメーションコネクタ付きTransformerアーキテクチャ" width="480" />
</td>
</tr>
<tr>
<td width="50%" valign="top">
<strong>GCPアーキテクチャ図</strong><br />
<p><strong>プロンプト:</strong> **GCPアイコン**を使用してGCPアーキテクチャ図を生成してください。この図では、ユーザーがインスタンス上でホストされているフロントエンドに接続します。</p>
<img src="../public/gcp_demo.svg" alt="GCPアーキテクチャ図" width="480" />
</td>
<td width="50%" valign="top">
<strong>AWSアーキテクチャ図</strong><br />
<p><strong>プロンプト:</strong> **AWSアイコン**を使用してAWSアーキテクチャ図を生成してください。この図では、ユーザーがインスタンス上でホストされているフロントエンドに接続します。</p>
<img src="../public/aws_demo.svg" alt="AWSアーキテクチャ図" width="480" />
</td>
</tr>
<tr>
<td width="50%" valign="top">
<strong>Azureアーキテクチャ図</strong><br />
<p><strong>プロンプト:</strong> **Azureアイコン**を使用してAzureアーキテクチャ図を生成してください。この図では、ユーザーがインスタンス上でホストされているフロントエンドに接続します。</p>
<img src="../public/azure_demo.svg" alt="Azureアーキテクチャ図" width="480" />
</td>
<td width="50%" valign="top">
<strong>猫のスケッチ</strong><br />
<p><strong>プロンプト:</strong> かわいい猫を描いてください。</p>
<img src="../public/cat_demo.svg" alt="猫の絵" width="240" />
</td>
</tr>
</table>
</div>
## 機能
- **LLM搭載のダイアグラム作成**大規模言語モデルを活用して、自然言語コマンドで直接draw.ioダイアグラムを作成・操作
- **画像ベースのダイアグラム複製**既存のダイアグラムや画像をアップロードし、AIが自動的に複製・強化
- **PDFとテキストファイルのアップロード**PDFドキュメントやテキストファイルをアップロードして、既存のドキュメントからコンテンツを抽出し、ダイアグラムを生成
- **AI推論プロセス表示**サポートされているモデルOpenAI o1/o3、Gemini、ClaudeなどのAIの思考プロセスを表示
- **ダイアグラム履歴**すべての変更を追跡する包括的なバージョン管理。AI編集前のダイアグラムの以前のバージョンを表示・復元可能
- **インタラクティブなチャットインターフェース**AIとリアルタイムでコミュニケーションしてダイアグラムを改善
- **クラウドアーキテクチャダイアグラムサポート**クラウドアーキテクチャダイアグラムの生成を専門的にサポートAWS、GCP、Azure
- **アニメーションコネクタ**:より良い可視化のためにダイアグラム要素間に動的でアニメーション化されたコネクタを作成
## はじめに
### オンラインで試す
インストール不要!デモサイトで直接お試しください:
[![Live Demo](../public/live-demo-button.svg)](https://next-ai-drawio.jiang.jp/)
> 注意:アクセス数が多いため、デモサイトでは現在 minimax-m2 モデルを使用しています。最高の結果を得るには、Claude Sonnet 4.5 または Claude Opus 4.5 でのセルフホスティングをお勧めします。
> **自分のAPIキーを使用**自分のAPIキーを使用することで、デモサイトの利用制限を回避できます。チャットパネルの設定アイコンをクリックして、プロバイダーとAPIキーを設定してください。キーはブラウザのローカルに保存され、サーバーには保存されません。
### Dockerで実行推奨
ローカルで実行したいだけなら、Dockerを使用するのが最も簡単です。
まず、Dockerをインストールしていない場合はインストールしてください[Dockerを入手](https://docs.docker.com/get-docker/)
次に実行:
```bash
docker run -d -p 3000:3000 \
-e AI_PROVIDER=openai \
-e AI_MODEL=gpt-4o \
-e OPENAI_API_KEY=your_api_key \
ghcr.io/dayuanjiang/next-ai-draw-io:latest
```
または env ファイルを使用:
```bash
cp env.example .env
# .env を編集して設定を入力
docker run -d -p 3000:3000 --env-file .env ghcr.io/dayuanjiang/next-ai-draw-io:latest
```
ブラウザで [http://localhost:3000](http://localhost:3000) を開いてください。
環境変数はお好みのAIプロバイダー設定に置き換えてください。利用可能なオプションについては[マルチプロバイダーサポート](#マルチプロバイダーサポート)を参照してください。
> **オフラインデプロイ:** `embed.diagrams.net` がブロックされている場合は、[オフラインデプロイガイド](./offline-deployment.md) で設定オプションをご確認ください。
### インストール
1. リポジトリをクローン:
```bash
git clone https://github.com/DayuanJiang/next-ai-draw-io
cd next-ai-draw-io
```
2. 依存関係をインストール:
```bash
npm install
```
3. AIプロバイダーを設定
ルートディレクトリに`.env.local`ファイルを作成:
```bash
cp env.example .env.local
```
`.env.local`を編集して選択したプロバイダーを設定:
- `AI_PROVIDER`を選択したプロバイダーに設定bedrock, openai, anthropic, google, azure, ollama, openrouter, deepseek, siliconflow
- `AI_MODEL`を使用する特定のモデルに設定
- プロバイダーに必要なAPIキーを追加
- `TEMPERATURE`:オプションの温度設定(例:`0`で決定論的な出力)。温度をサポートしないモデル(推論モデルなど)では設定しないでください。
- `ACCESS_CODE_LIST` アクセスパスワード(オプション)。カンマ区切りで複数のパスワードを指定できます。
> 警告:`ACCESS_CODE_LIST`を設定しない場合、誰でもデプロイされたサイトに直接アクセスできるため、トークンが急速に消費される可能性があります。このオプションを設定することをお勧めします。
詳細な設定手順については[プロバイダー設定ガイド](./ai-providers.md)を参照してください。
4. 開発サーバーを起動:
```bash
npm run dev
```
5. ブラウザで[http://localhost:3000](http://localhost:3000)を開いてアプリケーションを確認。
## デプロイ
Next.jsアプリをデプロイする最も簡単な方法は、Next.jsの作成者による[Vercelプラットフォーム](https://vercel.com/new)を使用することです。
詳細は[Next.jsデプロイメントドキュメント](https://nextjs.org/docs/app/building-your-application/deploying)をご覧ください。
または、このボタンでデプロイできます:
[![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https%3A%2F%2Fgithub.com%2FDayuanJiang%2Fnext-ai-draw-io)
ローカルの`.env.local`ファイルと同様に、Vercelダッシュボードで**環境変数を設定**してください。
## マルチプロバイダーサポート
- AWS Bedrockデフォルト
- OpenAI
- Anthropic
- Google AI
- Azure OpenAI
- Ollama
- OpenRouter
- DeepSeek
- SiliconFlow
AWS BedrockとOpenRouter以外のすべてのプロバイダーはカスタムエンドポイントをサポートしています。
📖 **[詳細なプロバイダー設定ガイド](./ai-providers.md)** - 各プロバイダーの設定手順をご覧ください。
**モデル要件**このタスクは厳密なフォーマット制約draw.io XMLを持つ長文テキスト生成を伴うため、強力なモデル機能が必要です。Claude Sonnet 4.5、GPT-4o、Gemini 2.0、DeepSeek V3/R1を推奨します。
注:`claude-sonnet-4-5`はAWSロゴ付きのdraw.ioダイアグラムで学習されているため、AWSアーキテクチャダイアグラムを作成したい場合は最適な選択です。
## 仕組み
本アプリケーションは以下の技術を使用しています:
- **Next.js**:フロントエンドフレームワークとルーティング
- **Vercel AI SDK**`ai` + `@ai-sdk/*`ストリーミングAIレスポンスとマルチプロバイダーサポート
- **react-drawio**:ダイアグラムの表現と操作
ダイアグラムはdraw.ioでレンダリングできるXMLとして表現されます。AIがコマンドを処理し、それに応じてこのXMLを生成または変更します。
## プロジェクト構造
```
app/ # Next.js App Router
api/chat/ # AIツール付きチャットAPIエンドポイント
page.tsx # DrawIO埋め込み付きメインページ
components/ # Reactコンポーネント
chat-panel.tsx # ダイアグラム制御付きチャットインターフェース
chat-input.tsx # ファイルアップロード付きユーザー入力コンポーネント
history-dialog.tsx # ダイアグラムバージョン履歴ビューア
ui/ # UIコンポーネントボタン、カードなど
contexts/ # Reactコンテキストプロバイダー
diagram-context.tsx # グローバルダイアグラム状態管理
lib/ # ユーティリティ関数とヘルパー
ai-providers.ts # マルチプロバイダーAI設定
utils.ts # XML処理と変換ユーティリティ
public/ # サンプル画像を含む静的アセット
```
## サポート&お問い合わせ
このプロジェクトが役に立ったら、ライブデモサイトのホスティングを支援するために[スポンサー](https://github.com/sponsors/DayuanJiang)をご検討ください!
サポートやお問い合わせについては、GitHubリポジトリでissueを開くか、メンテナーにご連絡ください
- メールme[at]jiang.jp
## スター履歴
[![Star History Chart](https://api.star-history.com/svg?repos=DayuanJiang/next-ai-draw-io&type=date&legend=top-left)](https://www.star-history.com/#DayuanJiang/next-ai-draw-io&type=date&legend=top-left)
---

178
docs/ai-providers.md Normal file
View File

@@ -0,0 +1,178 @@
# AI Provider Configuration
This guide explains how to configure different AI model providers for next-ai-draw-io.
## Quick Start
1. Copy `.env.example` to `.env.local`
2. Set your API key for your chosen provider
3. Set `AI_MODEL` to your desired model
4. Run `npm run dev`
## Supported Providers
### Google Gemini
```bash
GOOGLE_GENERATIVE_AI_API_KEY=your_api_key
AI_MODEL=gemini-2.0-flash
```
Optional custom endpoint:
```bash
GOOGLE_BASE_URL=https://your-custom-endpoint
```
### OpenAI
```bash
OPENAI_API_KEY=your_api_key
AI_MODEL=gpt-4o
```
Optional custom endpoint (for OpenAI-compatible services):
```bash
OPENAI_BASE_URL=https://your-custom-endpoint/v1
```
### Anthropic
```bash
ANTHROPIC_API_KEY=your_api_key
AI_MODEL=claude-sonnet-4-5-20250514
```
Optional custom endpoint:
```bash
ANTHROPIC_BASE_URL=https://your-custom-endpoint
```
### DeepSeek
```bash
DEEPSEEK_API_KEY=your_api_key
AI_MODEL=deepseek-chat
```
Optional custom endpoint:
```bash
DEEPSEEK_BASE_URL=https://your-custom-endpoint
```
### SiliconFlow (OpenAI-compatible)
```bash
SILICONFLOW_API_KEY=your_api_key
AI_MODEL=deepseek-ai/DeepSeek-V3 # example; use any SiliconFlow model id
```
Optional custom endpoint (defaults to the recommended domain):
```bash
SILICONFLOW_BASE_URL=https://api.siliconflow.com/v1 # or https://api.siliconflow.cn/v1
```
### Azure OpenAI
```bash
AZURE_API_KEY=your_api_key
AZURE_RESOURCE_NAME=your-resource-name # Required: your Azure resource name
AI_MODEL=your-deployment-name
```
Or use a custom endpoint instead of resource name:
```bash
AZURE_API_KEY=your_api_key
AZURE_BASE_URL=https://your-resource.openai.azure.com # Alternative to AZURE_RESOURCE_NAME
AI_MODEL=your-deployment-name
```
Optional reasoning configuration:
```bash
AZURE_REASONING_EFFORT=low # Optional: low, medium, high
AZURE_REASONING_SUMMARY=detailed # Optional: none, brief, detailed
```
### AWS Bedrock
```bash
AWS_REGION=us-west-2
AWS_ACCESS_KEY_ID=your_access_key_id
AWS_SECRET_ACCESS_KEY=your_secret_access_key
AI_MODEL=anthropic.claude-sonnet-4-5-20250514-v1:0
```
Note: On AWS (Lambda, EC2 with IAM role), credentials are automatically obtained from the IAM role.
### OpenRouter
```bash
OPENROUTER_API_KEY=your_api_key
AI_MODEL=anthropic/claude-sonnet-4
```
Optional custom endpoint:
```bash
OPENROUTER_BASE_URL=https://your-custom-endpoint
```
### Ollama (Local)
```bash
AI_PROVIDER=ollama
AI_MODEL=llama3.2
```
Optional custom URL:
```bash
OLLAMA_BASE_URL=http://localhost:11434
```
## Auto-Detection
If you only configure **one** provider's API key, the system will automatically detect and use that provider. No need to set `AI_PROVIDER`.
If you configure **multiple** API keys, you must explicitly set `AI_PROVIDER`:
```bash
AI_PROVIDER=google # or: openai, anthropic, deepseek, siliconflow, azure, bedrock, openrouter, ollama
```
## Model Capability Requirements
This task requires exceptionally strong model capabilities, as it involves generating long-form text with strict formatting constraints (draw.io XML).
**Recommended models**:
- Claude Sonnet 4.5 / Opus 4.5
**Note on Ollama**: While Ollama is supported as a provider, it's generally not practical for this use case unless you're running high-capability models like DeepSeek R1 or Qwen3-235B locally.
## Temperature Setting
You can optionally configure the temperature via environment variable:
```bash
TEMPERATURE=0 # More deterministic output (recommended for diagrams)
```
**Important**: Leave `TEMPERATURE` unset for models that don't support temperature settings, such as:
- GPT-5.1 and other reasoning models
- Some specialized models
When unset, the model uses its default behavior.
## Recommendations
- **Best experience**: Use models with vision support (GPT-4o, Claude, Gemini) for image-to-diagram features
- **Budget-friendly**: DeepSeek offers competitive pricing
- **Privacy**: Use Ollama for fully local, offline operation (requires powerful hardware)
- **Flexibility**: OpenRouter provides access to many models through a single API

View File

@@ -0,0 +1,39 @@
# Offline Deployment
Deploy Next AI Draw.io offline by self-hosting draw.io to replace `embed.diagrams.net`.
**Note:** `NEXT_PUBLIC_DRAWIO_BASE_URL` is a **build-time** variable. Changing it requires rebuilding the Docker image.
## Docker Compose Setup
1. Clone the repository and define API keys in `.env`.
2. Create `docker-compose.yml`:
```yaml
services:
drawio:
image: jgraph/drawio:latest
ports: ["8080:8080"]
next-ai-draw-io:
build:
context: .
args:
- NEXT_PUBLIC_DRAWIO_BASE_URL=http://localhost:8080
ports: ["3000:3000"]
env_file: .env
depends_on: [drawio]
```
3. Run `docker compose up -d` and open `http://localhost:3000`.
## Configuration & Critical Warning
**The `NEXT_PUBLIC_DRAWIO_BASE_URL` must be accessible from the user's browser.**
| Scenario | URL Value |
|----------|-----------|
| Localhost | `http://localhost:8080` |
| Remote/Server | `http://YOUR_SERVER_IP:8080` or `https://drawio.your-domain.com` |
**Do NOT use** internal Docker aliases like `http://drawio:8080`; the browser cannot resolve them.

View File

@@ -1,6 +1,6 @@
# AI Provider Configuration
# AI_PROVIDER: Which provider to use
# Options: bedrock, openai, anthropic, google, azure, ollama, openrouter
# Options: bedrock, openai, anthropic, google, azure, ollama, openrouter, deepseek, siliconflow
# Default: bedrock
AI_PROVIDER=bedrock
@@ -11,25 +11,83 @@ AI_MODEL=global.anthropic.claude-sonnet-4-5-20250929-v1:0
# AWS_REGION=us-east-1
# AWS_ACCESS_KEY_ID=your-access-key-id
# AWS_SECRET_ACCESS_KEY=your-secret-access-key
# Note: Claude and Nova models support reasoning/extended thinking
# BEDROCK_REASONING_BUDGET_TOKENS=12000 # Optional: Claude reasoning budget in tokens (1024-64000)
# BEDROCK_REASONING_EFFORT=medium # Optional: Nova reasoning effort (low/medium/high)
# OpenAI Configuration
# OPENAI_API_KEY=sk-...
# OPENAI_BASE_URL=https://api.openai.com/v1 # Optional: Custom OpenAI-compatible endpoint
# OPENAI_ORGANIZATION=org-... # Optional
# OPENAI_PROJECT=proj_... # Optional
# Note: o1/o3/gpt-5 models automatically enable reasoning summary (default: detailed)
# OPENAI_REASONING_EFFORT=low # Optional: Reasoning effort (minimal/low/medium/high) - for o1/o3/gpt-5
# OPENAI_REASONING_SUMMARY=detailed # Optional: Override reasoning summary (none/brief/detailed)
# Anthropic (Direct) Configuration
# ANTHROPIC_API_KEY=sk-ant-...
# ANTHROPIC_BASE_URL=https://your-custom-anthropic/v1
# ANTHROPIC_THINKING_TYPE=enabled # Optional: Anthropic extended thinking (enabled)
# ANTHROPIC_THINKING_BUDGET_TOKENS=12000 # Optional: Budget for extended thinking in tokens
# Google Generative AI Configuration
# GOOGLE_GENERATIVE_AI_API_KEY=...
# GOOGLE_BASE_URL=https://generativelanguage.googleapis.com/v1beta # Optional: Custom endpoint
# GOOGLE_CANDIDATE_COUNT=1 # Optional: Number of candidates to generate
# GOOGLE_TOP_K=40 # Optional: Top K sampling parameter
# GOOGLE_TOP_P=0.95 # Optional: Nucleus sampling parameter
# Note: Gemini 2.5/3 models automatically enable reasoning display (includeThoughts: true)
# GOOGLE_THINKING_BUDGET=8192 # Optional: Gemini 2.5 thinking budget in tokens (for more/less thinking)
# GOOGLE_THINKING_LEVEL=high # Optional: Gemini 3 thinking level (low/high)
# Azure OpenAI Configuration
# Configure endpoint using ONE of these methods:
# 1. AZURE_RESOURCE_NAME - SDK constructs: https://{name}.openai.azure.com/openai/v1{path}
# 2. AZURE_BASE_URL - SDK appends /v1{path} to your URL
# If both are set, AZURE_BASE_URL takes precedence.
# AZURE_RESOURCE_NAME=your-resource-name
# AZURE_API_KEY=...
# AZURE_BASE_URL=https://your-resource.openai.azure.com/openai # Alternative: Custom endpoint
# AZURE_REASONING_EFFORT=low # Optional: Azure reasoning effort (low, medium, high)
# AZURE_REASONING_SUMMARY=detailed
# Ollama (Local) Configuration
# OLLAMA_BASE_URL=http://localhost:11434/api # Optional, defaults to localhost
# OLLAMA_ENABLE_THINKING=true # Optional: Enable thinking for models that support it (e.g., qwen3)
# OpenRouter Configuration
# OPENROUTER_API_KEY=sk-or-v1-...
# OPENROUTER_BASE_URL=https://openrouter.ai/api/v1 # Optional: Custom endpoint
# DeepSeek Configuration
# DEEPSEEK_API_KEY=sk-...
# DEEPSEEK_BASE_URL=https://api.deepseek.com/v1 # Optional: Custom endpoint
# SiliconFlow Configuration (OpenAI-compatible)
# Base domain can be .com or .cn, defaults to https://api.siliconflow.com/v1
# SILICONFLOW_API_KEY=sk-...
# SILICONFLOW_BASE_URL=https://api.siliconflow.com/v1 # Optional: switch to https://api.siliconflow.cn/v1 if needed
# Langfuse Observability (Optional)
# Enable LLM tracing and analytics - https://langfuse.com
# LANGFUSE_PUBLIC_KEY=pk-lf-...
# LANGFUSE_SECRET_KEY=sk-lf-...
# LANGFUSE_BASEURL=https://cloud.langfuse.com # EU region, use https://us.cloud.langfuse.com for US
# Temperature (Optional)
# Controls randomness in AI responses. Lower = more deterministic.
# Leave unset for models that don't support temperature (e.g., GPT-5.1 reasoning models)
# TEMPERATURE=0
# Access Control (Optional)
# ACCESS_CODE_LIST=your-secret-code,another-code
# Draw.io Configuration (Optional)
# NEXT_PUBLIC_DRAWIO_BASE_URL=https://embed.diagrams.net # Default: https://embed.diagrams.net
# Use this to point to a self-hosted draw.io instance
# PDF Input Feature (Optional)
# Enable PDF file upload to extract text and generate diagrams
# Enabled by default. Set to "false" to disable.
# ENABLE_PDF_INPUT=true
# NEXT_PUBLIC_MAX_EXTRACTED_CHARS=150000 # Max characters for PDF/text extraction (default: 150000)

39
instrumentation.ts Normal file
View File

@@ -0,0 +1,39 @@
import { LangfuseSpanProcessor } from "@langfuse/otel"
import { NodeTracerProvider } from "@opentelemetry/sdk-trace-node"
export function register() {
// Skip telemetry if Langfuse env vars are not configured
if (!process.env.LANGFUSE_PUBLIC_KEY || !process.env.LANGFUSE_SECRET_KEY) {
console.warn(
"[Langfuse] Environment variables not configured - telemetry disabled",
)
return
}
const langfuseSpanProcessor = new LangfuseSpanProcessor({
publicKey: process.env.LANGFUSE_PUBLIC_KEY,
secretKey: process.env.LANGFUSE_SECRET_KEY,
baseUrl: process.env.LANGFUSE_BASEURL,
// Filter out Next.js HTTP request spans so AI SDK spans become root traces
shouldExportSpan: ({ otelSpan }) => {
const spanName = otelSpan.name
// Skip Next.js HTTP infrastructure spans
if (
spanName.startsWith("POST /") ||
spanName.startsWith("GET /") ||
spanName.includes("BaseServer") ||
spanName.includes("handleRequest")
) {
return false
}
return true
},
})
const tracerProvider = new NodeTracerProvider({
spanProcessors: [langfuseSpanProcessor],
})
// Register globally so AI SDK's telemetry also uses this processor
tracerProvider.register()
}

26
lib/ai-config.ts Normal file
View File

@@ -0,0 +1,26 @@
import { STORAGE_KEYS } from "./storage"
/**
* Get AI configuration from localStorage.
* Returns API keys and settings for custom AI providers.
* Used to override server defaults when user provides their own API key.
*/
export function getAIConfig() {
if (typeof window === "undefined") {
return {
accessCode: "",
aiProvider: "",
aiBaseUrl: "",
aiApiKey: "",
aiModel: "",
}
}
return {
accessCode: localStorage.getItem(STORAGE_KEYS.accessCode) || "",
aiProvider: localStorage.getItem(STORAGE_KEYS.aiProvider) || "",
aiBaseUrl: localStorage.getItem(STORAGE_KEYS.aiBaseUrl) || "",
aiApiKey: localStorage.getItem(STORAGE_KEYS.aiApiKey) || "",
aiModel: localStorage.getItem(STORAGE_KEYS.aiModel) || "",
}
}

View File

@@ -1,62 +1,426 @@
import { bedrock } from '@ai-sdk/amazon-bedrock';
import { openai, createOpenAI } from '@ai-sdk/openai';
import { anthropic } from '@ai-sdk/anthropic';
import { google } from '@ai-sdk/google';
import { azure } from '@ai-sdk/azure';
import { ollama } from 'ollama-ai-provider-v2';
import { createOpenRouter } from '@openrouter/ai-sdk-provider';
import { createAmazonBedrock } from "@ai-sdk/amazon-bedrock"
import { createAnthropic } from "@ai-sdk/anthropic"
import { azure, createAzure } from "@ai-sdk/azure"
import { createDeepSeek, deepseek } from "@ai-sdk/deepseek"
import { createGoogleGenerativeAI, google } from "@ai-sdk/google"
import { createOpenAI, openai } from "@ai-sdk/openai"
import { fromNodeProviderChain } from "@aws-sdk/credential-providers"
import { createOpenRouter } from "@openrouter/ai-sdk-provider"
import { createOllama, ollama } from "ollama-ai-provider-v2"
export type ProviderName =
| 'bedrock'
| 'openai'
| 'anthropic'
| 'google'
| 'azure'
| 'ollama'
| 'openrouter';
| "bedrock"
| "openai"
| "anthropic"
| "google"
| "azure"
| "ollama"
| "openrouter"
| "deepseek"
| "siliconflow"
interface ModelConfig {
model: any;
providerOptions?: any;
model: any
providerOptions?: any
headers?: Record<string, string>
modelId: string
}
// Anthropic beta headers for fine-grained tool streaming
const ANTHROPIC_BETA_OPTIONS = {
anthropic: {
additionalModelRequestFields: {
anthropic_beta: ['fine-grained-tool-streaming-2025-05-14']
export interface ClientOverrides {
provider?: string | null
baseUrl?: string | null
apiKey?: string | null
modelId?: string | null
}
// Providers that can be used with client-provided API keys
const ALLOWED_CLIENT_PROVIDERS: ProviderName[] = [
"openai",
"anthropic",
"google",
"azure",
"openrouter",
"deepseek",
"siliconflow",
]
// Bedrock provider options for Anthropic beta features
const BEDROCK_ANTHROPIC_BETA = {
bedrock: {
anthropicBeta: ["fine-grained-tool-streaming-2025-05-14"],
},
}
// Direct Anthropic API headers for beta features
const ANTHROPIC_BETA_HEADERS = {
"anthropic-beta": "fine-grained-tool-streaming-2025-05-14",
}
/**
* Safely parse integer from environment variable with validation
*/
function parseIntSafe(
value: string | undefined,
varName: string,
min?: number,
max?: number,
): number | undefined {
if (!value) return undefined
const parsed = Number.parseInt(value, 10)
if (Number.isNaN(parsed)) {
throw new Error(`${varName} must be a valid integer, got: ${value}`)
}
}
};
if (min !== undefined && parsed < min) {
throw new Error(`${varName} must be >= ${min}, got: ${parsed}`)
}
if (max !== undefined && parsed > max) {
throw new Error(`${varName} must be <= ${max}, got: ${parsed}`)
}
return parsed
}
/**
* Build provider-specific options from environment variables
* Supports various AI SDK providers with their unique configuration options
*
* Environment variables:
* - OPENAI_REASONING_EFFORT: OpenAI reasoning effort level (minimal/low/medium/high) - for o1/o3/gpt-5
* - OPENAI_REASONING_SUMMARY: OpenAI reasoning summary (none/brief/detailed) - auto-enabled for o1/o3/gpt-5
* - ANTHROPIC_THINKING_BUDGET_TOKENS: Anthropic thinking budget in tokens (1024-64000)
* - ANTHROPIC_THINKING_TYPE: Anthropic thinking type (enabled)
* - GOOGLE_THINKING_BUDGET: Google Gemini 2.5 thinking budget in tokens (1024-100000)
* - GOOGLE_THINKING_LEVEL: Google Gemini 3 thinking level (low/high)
* - AZURE_REASONING_EFFORT: Azure/OpenAI reasoning effort (low/medium/high)
* - AZURE_REASONING_SUMMARY: Azure reasoning summary (none/brief/detailed)
* - BEDROCK_REASONING_BUDGET_TOKENS: Bedrock Claude reasoning budget in tokens (1024-64000)
* - BEDROCK_REASONING_EFFORT: Bedrock Nova reasoning effort (low/medium/high)
* - OLLAMA_ENABLE_THINKING: Enable Ollama thinking mode (set to "true")
*/
function buildProviderOptions(
provider: ProviderName,
modelId?: string,
): Record<string, any> | undefined {
const options: Record<string, any> = {}
switch (provider) {
case "openai": {
const reasoningEffort = process.env.OPENAI_REASONING_EFFORT
const reasoningSummary = process.env.OPENAI_REASONING_SUMMARY
// OpenAI reasoning models (o1, o3, gpt-5) need reasoningSummary to return thoughts
if (
modelId &&
(modelId.includes("o1") ||
modelId.includes("o3") ||
modelId.includes("gpt-5"))
) {
options.openai = {
// Auto-enable reasoning summary for reasoning models (default: detailed)
reasoningSummary:
(reasoningSummary as "none" | "brief" | "detailed") ||
"detailed",
}
// Optionally configure reasoning effort
if (reasoningEffort) {
options.openai.reasoningEffort = reasoningEffort as
| "minimal"
| "low"
| "medium"
| "high"
}
} else if (reasoningEffort || reasoningSummary) {
// Non-reasoning models: only apply if explicitly configured
options.openai = {}
if (reasoningEffort) {
options.openai.reasoningEffort = reasoningEffort as
| "minimal"
| "low"
| "medium"
| "high"
}
if (reasoningSummary) {
options.openai.reasoningSummary = reasoningSummary as
| "none"
| "brief"
| "detailed"
}
}
break
}
case "anthropic": {
const thinkingBudget = parseIntSafe(
process.env.ANTHROPIC_THINKING_BUDGET_TOKENS,
"ANTHROPIC_THINKING_BUDGET_TOKENS",
1024,
64000,
)
const thinkingType =
process.env.ANTHROPIC_THINKING_TYPE || "enabled"
if (thinkingBudget) {
options.anthropic = {
thinking: {
type: thinkingType,
budgetTokens: thinkingBudget,
},
}
}
break
}
case "google": {
const reasoningEffort = process.env.GOOGLE_REASONING_EFFORT
const thinkingBudgetVal = parseIntSafe(
process.env.GOOGLE_THINKING_BUDGET,
"GOOGLE_THINKING_BUDGET",
1024,
100000,
)
const thinkingLevel = process.env.GOOGLE_THINKING_LEVEL
// Google Gemini 2.5/3 models think by default, but need includeThoughts: true
// to return the reasoning in the response
if (
modelId &&
(modelId.includes("gemini-2") ||
modelId.includes("gemini-3") ||
modelId.includes("gemini2") ||
modelId.includes("gemini3"))
) {
const thinkingConfig: Record<string, any> = {
includeThoughts: true,
}
// Optionally configure thinking budget or level
if (
thinkingBudgetVal &&
(modelId.includes("2.5") || modelId.includes("2-5"))
) {
thinkingConfig.thinkingBudget = thinkingBudgetVal
} else if (
thinkingLevel &&
(modelId.includes("gemini-3") ||
modelId.includes("gemini3"))
) {
thinkingConfig.thinkingLevel = thinkingLevel as
| "low"
| "high"
}
options.google = { thinkingConfig }
} else if (reasoningEffort) {
options.google = {
reasoningEffort: reasoningEffort as
| "low"
| "medium"
| "high",
}
}
// Keep existing Google options
const options_obj: Record<string, any> = {}
const candidateCount = parseIntSafe(
process.env.GOOGLE_CANDIDATE_COUNT,
"GOOGLE_CANDIDATE_COUNT",
1,
8,
)
if (candidateCount) {
options_obj.candidateCount = candidateCount
}
const topK = parseIntSafe(
process.env.GOOGLE_TOP_K,
"GOOGLE_TOP_K",
1,
100,
)
if (topK) {
options_obj.topK = topK
}
if (process.env.GOOGLE_TOP_P) {
const topP = Number.parseFloat(process.env.GOOGLE_TOP_P)
if (Number.isNaN(topP) || topP < 0 || topP > 1) {
throw new Error(
`GOOGLE_TOP_P must be a number between 0 and 1, got: ${process.env.GOOGLE_TOP_P}`,
)
}
options_obj.topP = topP
}
if (Object.keys(options_obj).length > 0) {
options.google = { ...options.google, ...options_obj }
}
break
}
case "azure": {
const reasoningEffort = process.env.AZURE_REASONING_EFFORT
const reasoningSummary = process.env.AZURE_REASONING_SUMMARY
if (reasoningEffort || reasoningSummary) {
options.azure = {}
if (reasoningEffort) {
options.azure.reasoningEffort = reasoningEffort as
| "low"
| "medium"
| "high"
}
if (reasoningSummary) {
options.azure.reasoningSummary = reasoningSummary as
| "none"
| "brief"
| "detailed"
}
}
break
}
case "bedrock": {
const budgetTokens = parseIntSafe(
process.env.BEDROCK_REASONING_BUDGET_TOKENS,
"BEDROCK_REASONING_BUDGET_TOKENS",
1024,
64000,
)
const reasoningEffort = process.env.BEDROCK_REASONING_EFFORT
// Bedrock reasoning ONLY for Claude and Nova models
// Other models (MiniMax, etc.) don't support reasoningConfig
if (
modelId &&
(budgetTokens || reasoningEffort) &&
(modelId.includes("claude") ||
modelId.includes("anthropic") ||
modelId.includes("nova") ||
modelId.includes("amazon"))
) {
const reasoningConfig: Record<string, any> = { type: "enabled" }
// Claude models: use budgetTokens (1024-64000)
if (
budgetTokens &&
(modelId.includes("claude") ||
modelId.includes("anthropic"))
) {
reasoningConfig.budgetTokens = budgetTokens
}
// Nova models: use maxReasoningEffort (low/medium/high)
else if (
reasoningEffort &&
(modelId.includes("nova") || modelId.includes("amazon"))
) {
reasoningConfig.maxReasoningEffort = reasoningEffort as
| "low"
| "medium"
| "high"
}
options.bedrock = { reasoningConfig }
}
break
}
case "ollama": {
const enableThinking = process.env.OLLAMA_ENABLE_THINKING
// Ollama supports reasoning with think: true for models like qwen3
if (enableThinking === "true") {
options.ollama = { think: true }
}
break
}
case "deepseek":
case "openrouter":
case "siliconflow": {
// These providers don't have reasoning configs in AI SDK yet
break
}
default:
break
}
return Object.keys(options).length > 0 ? options : undefined
}
// Map of provider to required environment variable
const PROVIDER_ENV_VARS: Record<ProviderName, string | null> = {
bedrock: null, // AWS SDK auto-uses IAM role on AWS, or env vars locally
openai: "OPENAI_API_KEY",
anthropic: "ANTHROPIC_API_KEY",
google: "GOOGLE_GENERATIVE_AI_API_KEY",
azure: "AZURE_API_KEY",
ollama: null, // No credentials needed for local Ollama
openrouter: "OPENROUTER_API_KEY",
deepseek: "DEEPSEEK_API_KEY",
siliconflow: "SILICONFLOW_API_KEY",
}
/**
* Auto-detect provider based on available API keys
* Returns the provider if exactly one is configured, otherwise null
*/
function detectProvider(): ProviderName | null {
const configuredProviders: ProviderName[] = []
for (const [provider, envVar] of Object.entries(PROVIDER_ENV_VARS)) {
if (envVar === null) {
// Skip ollama - it doesn't require credentials
continue
}
if (process.env[envVar]) {
// Azure requires additional config (baseURL or resourceName)
if (provider === "azure") {
const hasBaseUrl = !!process.env.AZURE_BASE_URL
const hasResourceName = !!process.env.AZURE_RESOURCE_NAME
if (hasBaseUrl || hasResourceName) {
configuredProviders.push(provider as ProviderName)
}
} else {
configuredProviders.push(provider as ProviderName)
}
}
}
if (configuredProviders.length === 1) {
return configuredProviders[0]
}
return null
}
/**
* Validate that required API keys are present for the selected provider
*/
function validateProviderCredentials(provider: ProviderName): void {
const requiredEnvVars: Record<ProviderName, string | null> = {
bedrock: 'AWS_ACCESS_KEY_ID',
openai: 'OPENAI_API_KEY',
anthropic: 'ANTHROPIC_API_KEY',
google: 'GOOGLE_GENERATIVE_AI_API_KEY',
azure: 'AZURE_API_KEY',
ollama: null, // No credentials needed for local Ollama
openrouter: 'OPENROUTER_API_KEY',
};
const requiredVar = PROVIDER_ENV_VARS[provider]
if (requiredVar && !process.env[requiredVar]) {
throw new Error(
`${requiredVar} environment variable is required for ${provider} provider. ` +
`Please set it in your .env.local file.`,
)
}
const requiredVar = requiredEnvVars[provider];
if (requiredVar && !process.env[requiredVar]) {
throw new Error(
`${requiredVar} environment variable is required for ${provider} provider. ` +
`Please set it in your .env.local file.`
);
}
// Azure requires either AZURE_BASE_URL or AZURE_RESOURCE_NAME in addition to API key
if (provider === "azure") {
const hasBaseUrl = !!process.env.AZURE_BASE_URL
const hasResourceName = !!process.env.AZURE_RESOURCE_NAME
if (!hasBaseUrl && !hasResourceName) {
throw new Error(
`Azure requires either AZURE_BASE_URL or AZURE_RESOURCE_NAME to be set. ` +
`Please set one in your .env.local file.`,
)
}
}
}
/**
* Get the AI model based on environment variables
*
* Environment variables:
* - AI_PROVIDER: The provider to use (bedrock, openai, anthropic, google, azure, ollama, openrouter)
* - AI_PROVIDER: The provider to use (bedrock, openai, anthropic, google, azure, ollama, openrouter, deepseek, siliconflow)
* - AI_MODEL: The model ID/name for the selected provider
*
* Provider-specific env vars:
@@ -68,82 +432,260 @@ function validateProviderCredentials(provider: ProviderName): void {
* - AWS_REGION, AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY: AWS Bedrock credentials
* - OLLAMA_BASE_URL: Ollama server URL (optional, defaults to http://localhost:11434)
* - OPENROUTER_API_KEY: OpenRouter API key
* - DEEPSEEK_API_KEY: DeepSeek API key
* - DEEPSEEK_BASE_URL: DeepSeek endpoint (optional)
* - SILICONFLOW_API_KEY: SiliconFlow API key
* - SILICONFLOW_BASE_URL: SiliconFlow endpoint (optional, defaults to https://api.siliconflow.com/v1)
*/
export function getAIModel(): ModelConfig {
const provider = (process.env.AI_PROVIDER || 'bedrock') as ProviderName;
const modelId = process.env.AI_MODEL;
export function getAIModel(overrides?: ClientOverrides): ModelConfig {
// Check if client is providing their own provider override
const isClientOverride = !!(overrides?.provider && overrides?.apiKey)
if (!modelId) {
throw new Error(
`AI_MODEL environment variable is required. Example: AI_MODEL=claude-sonnet-4-5`
);
}
// Use client override if provided, otherwise fall back to env vars
const modelId = overrides?.modelId || process.env.AI_MODEL
// Validate provider credentials
validateProviderCredentials(provider);
if (!modelId) {
if (isClientOverride) {
throw new Error(
`Model ID is required when using custom AI provider. Please specify a model in Settings.`,
)
}
throw new Error(
`AI_MODEL environment variable is required. Example: AI_MODEL=claude-sonnet-4-5`,
)
}
// Log initialization for debugging
console.log(`[AI Provider] Initializing ${provider} with model: ${modelId}`);
// Determine provider: client override > explicit config > auto-detect > error
let provider: ProviderName
if (overrides?.provider) {
// Validate client-provided provider
if (
!ALLOWED_CLIENT_PROVIDERS.includes(
overrides.provider as ProviderName,
)
) {
throw new Error(
`Invalid provider: ${overrides.provider}. Allowed providers: ${ALLOWED_CLIENT_PROVIDERS.join(", ")}`,
)
}
provider = overrides.provider as ProviderName
} else if (process.env.AI_PROVIDER) {
provider = process.env.AI_PROVIDER as ProviderName
} else {
const detected = detectProvider()
if (detected) {
provider = detected
console.log(`[AI Provider] Auto-detected provider: ${provider}`)
} else {
// List configured providers for better error message
const configured = Object.entries(PROVIDER_ENV_VARS)
.filter(([, envVar]) => envVar && process.env[envVar as string])
.map(([p]) => p)
let model: any;
let providerOptions: any = undefined;
if (configured.length === 0) {
throw new Error(
`No AI provider configured. Please set one of the following API keys in your .env.local file:\n` +
`- DEEPSEEK_API_KEY for DeepSeek\n` +
`- OPENAI_API_KEY for OpenAI\n` +
`- ANTHROPIC_API_KEY for Anthropic\n` +
`- GOOGLE_GENERATIVE_AI_API_KEY for Google\n` +
`- AWS_ACCESS_KEY_ID for Bedrock\n` +
`- OPENROUTER_API_KEY for OpenRouter\n` +
`- AZURE_API_KEY for Azure\n` +
`- SILICONFLOW_API_KEY for SiliconFlow\n` +
`Or set AI_PROVIDER=ollama for local Ollama.`,
)
} else {
throw new Error(
`Multiple AI providers configured (${configured.join(", ")}). ` +
`Please set AI_PROVIDER to specify which one to use.`,
)
}
}
}
switch (provider) {
case 'bedrock':
model = bedrock(modelId);
// Add Anthropic beta headers if using Claude models via Bedrock
if (modelId.includes('anthropic.claude')) {
providerOptions = ANTHROPIC_BETA_OPTIONS;
}
break;
// Only validate server credentials if client isn't providing their own API key
if (!isClientOverride) {
validateProviderCredentials(provider)
}
case 'openai':
if (process.env.OPENAI_BASE_URL) {
const customOpenAI = createOpenAI({
apiKey: process.env.OPENAI_API_KEY,
baseURL: process.env.OPENAI_BASE_URL,
});
model = customOpenAI.chat(modelId);
} else {
model = openai(modelId);
}
break;
console.log(`[AI Provider] Initializing ${provider} with model: ${modelId}`)
case 'anthropic':
model = anthropic(modelId);
// Add beta headers for fine-grained tool streaming
providerOptions = ANTHROPIC_BETA_OPTIONS;
break;
let model: any
let providerOptions: any
let headers: Record<string, string> | undefined
case 'google':
model = google(modelId);
break;
// Build provider-specific options from environment variables
const customProviderOptions = buildProviderOptions(provider, modelId)
case 'azure':
model = azure(modelId);
break;
switch (provider) {
case "bedrock": {
// Use credential provider chain for IAM role support (Lambda, EC2, etc.)
// Falls back to env vars (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY) for local dev
const bedrockProvider = createAmazonBedrock({
region: process.env.AWS_REGION || "us-west-2",
credentialProvider: fromNodeProviderChain(),
})
model = bedrockProvider(modelId)
// Add Anthropic beta options if using Claude models via Bedrock
if (modelId.includes("anthropic.claude")) {
// Deep merge to preserve both anthropicBeta and reasoningConfig
providerOptions = {
bedrock: {
...BEDROCK_ANTHROPIC_BETA.bedrock,
...(customProviderOptions?.bedrock || {}),
},
}
} else if (customProviderOptions) {
providerOptions = customProviderOptions
}
break
}
case 'ollama':
model = ollama(modelId);
break;
case "openai": {
const apiKey = overrides?.apiKey || process.env.OPENAI_API_KEY
const baseURL = overrides?.baseUrl || process.env.OPENAI_BASE_URL
if (baseURL || overrides?.apiKey) {
const customOpenAI = createOpenAI({
apiKey,
...(baseURL && { baseURL }),
})
model = customOpenAI.chat(modelId)
} else {
model = openai(modelId)
}
break
}
case 'openrouter':
const openrouter = createOpenRouter({
apiKey: process.env.OPENROUTER_API_KEY,
});
model = openrouter(modelId);
break;
case "anthropic": {
const apiKey = overrides?.apiKey || process.env.ANTHROPIC_API_KEY
const baseURL =
overrides?.baseUrl ||
process.env.ANTHROPIC_BASE_URL ||
"https://api.anthropic.com/v1"
const customProvider = createAnthropic({
apiKey,
baseURL,
headers: ANTHROPIC_BETA_HEADERS,
})
model = customProvider(modelId)
// Add beta headers for fine-grained tool streaming
headers = ANTHROPIC_BETA_HEADERS
break
}
default:
throw new Error(
`Unknown AI provider: ${provider}. Supported providers: bedrock, openai, anthropic, google, azure, ollama, openrouter`
);
}
case "google": {
const apiKey =
overrides?.apiKey || process.env.GOOGLE_GENERATIVE_AI_API_KEY
const baseURL = overrides?.baseUrl || process.env.GOOGLE_BASE_URL
if (baseURL || overrides?.apiKey) {
const customGoogle = createGoogleGenerativeAI({
apiKey,
...(baseURL && { baseURL }),
})
model = customGoogle(modelId)
} else {
model = google(modelId)
}
break
}
// Log if provider options are being applied
if (providerOptions) {
console.log('[AI Provider] Applying provider-specific options');
}
case "azure": {
const apiKey = overrides?.apiKey || process.env.AZURE_API_KEY
const baseURL = overrides?.baseUrl || process.env.AZURE_BASE_URL
const resourceName = process.env.AZURE_RESOURCE_NAME
// Azure requires either baseURL or resourceName to construct the endpoint
// resourceName constructs: https://{resourceName}.openai.azure.com/openai/v1{path}
if (baseURL || resourceName || overrides?.apiKey) {
const customAzure = createAzure({
apiKey,
// baseURL takes precedence over resourceName per SDK behavior
...(baseURL && { baseURL }),
...(!baseURL && resourceName && { resourceName }),
})
model = customAzure(modelId)
} else {
model = azure(modelId)
}
break
}
return { model, providerOptions };
case "ollama":
if (process.env.OLLAMA_BASE_URL) {
const customOllama = createOllama({
baseURL: process.env.OLLAMA_BASE_URL,
})
model = customOllama(modelId)
} else {
model = ollama(modelId)
}
break
case "openrouter": {
const apiKey = overrides?.apiKey || process.env.OPENROUTER_API_KEY
const baseURL =
overrides?.baseUrl || process.env.OPENROUTER_BASE_URL
const openrouter = createOpenRouter({
apiKey,
...(baseURL && { baseURL }),
})
model = openrouter(modelId)
break
}
case "deepseek": {
const apiKey = overrides?.apiKey || process.env.DEEPSEEK_API_KEY
const baseURL = overrides?.baseUrl || process.env.DEEPSEEK_BASE_URL
if (baseURL || overrides?.apiKey) {
const customDeepSeek = createDeepSeek({
apiKey,
...(baseURL && { baseURL }),
})
model = customDeepSeek(modelId)
} else {
model = deepseek(modelId)
}
break
}
case "siliconflow": {
const apiKey = overrides?.apiKey || process.env.SILICONFLOW_API_KEY
const baseURL =
overrides?.baseUrl ||
process.env.SILICONFLOW_BASE_URL ||
"https://api.siliconflow.com/v1"
const siliconflowProvider = createOpenAI({
apiKey,
baseURL,
})
model = siliconflowProvider.chat(modelId)
break
}
default:
throw new Error(
`Unknown AI provider: ${provider}. Supported providers: bedrock, openai, anthropic, google, azure, ollama, openrouter, deepseek, siliconflow`,
)
}
// Apply provider-specific options for all providers except bedrock (which has special handling)
if (customProviderOptions && provider !== "bedrock" && !providerOptions) {
providerOptions = customProviderOptions
}
return { model, providerOptions, headers, modelId }
}
/**
* Check if a model supports prompt caching.
* Currently only Claude models on Bedrock support prompt caching.
*/
export function supportsPromptCaching(modelId: string): boolean {
// Bedrock prompt caching is supported for Claude models
return (
modelId.includes("claude") ||
modelId.includes("anthropic") ||
modelId.startsWith("us.anthropic") ||
modelId.startsWith("eu.anthropic")
)
}

892
lib/cached-responses.ts Normal file
View File

@@ -0,0 +1,892 @@
export interface CachedResponse {
promptText: string
hasImage: boolean
xml: string
}
export const CACHED_EXAMPLE_RESPONSES: CachedResponse[] = [
{
promptText:
"Give me a **animated connector** diagram of transformer's architecture",
hasImage: false,
xml: `<mxCell id="title" value="Transformer Architecture" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=20;fontStyle=1;" vertex="1" parent="1">
<mxGeometry x="300" y="20" width="250" height="30" as="geometry"/>
</mxCell>
<mxCell id="input_embed" value="Input Embedding" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#dae8fc;strokeColor=#6c8ebf;fontSize=11;" vertex="1" parent="1">
<mxGeometry x="80" y="480" width="120" height="40" as="geometry"/>
</mxCell>
<mxCell id="pos_enc_left" value="Positional Encoding" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#dae8fc;strokeColor=#6c8ebf;fontSize=11;" vertex="1" parent="1">
<mxGeometry x="80" y="420" width="120" height="40" as="geometry"/>
</mxCell>
<mxCell id="encoder_box" value="ENCODER" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#e1d5e7;strokeColor=#9673a6;verticalAlign=top;fontSize=12;fontStyle=1;" vertex="1" parent="1">
<mxGeometry x="60" y="180" width="160" height="220" as="geometry"/>
</mxCell>
<mxCell id="mha_enc" value="Multi-Head&#xa;Attention" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#fff2cc;strokeColor=#d6b656;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="80" y="330" width="120" height="50" as="geometry"/>
</mxCell>
<mxCell id="add_norm1_enc" value="Add &amp; Norm" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#d5e8d4;strokeColor=#82b366;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="80" y="280" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="ff_enc" value="Feed Forward" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#fff2cc;strokeColor=#d6b656;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="80" y="240" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="add_norm2_enc" value="Add &amp; Norm" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#d5e8d4;strokeColor=#82b366;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="80" y="200" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="nx_enc" value="Nx" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;fontStyle=2;" vertex="1" parent="1">
<mxGeometry x="30" y="275" width="30" height="30" as="geometry"/>
</mxCell>
<mxCell id="output_embed" value="Output Embedding" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#dae8fc;strokeColor=#6c8ebf;fontSize=11;" vertex="1" parent="1">
<mxGeometry x="650" y="480" width="120" height="40" as="geometry"/>
</mxCell>
<mxCell id="pos_enc_right" value="Positional Encoding" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#dae8fc;strokeColor=#6c8ebf;fontSize=11;" vertex="1" parent="1">
<mxGeometry x="650" y="420" width="120" height="40" as="geometry"/>
</mxCell>
<mxCell id="decoder_box" value="DECODER" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffe6cc;strokeColor=#d79b00;verticalAlign=top;fontSize=12;fontStyle=1;" vertex="1" parent="1">
<mxGeometry x="630" y="140" width="160" height="260" as="geometry"/>
</mxCell>
<mxCell id="masked_mha_dec" value="Masked Multi-Head&#xa;Attention" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#fff2cc;strokeColor=#d6b656;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="650" y="340" width="120" height="50" as="geometry"/>
</mxCell>
<mxCell id="add_norm1_dec" value="Add &amp; Norm" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#d5e8d4;strokeColor=#82b366;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="650" y="290" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="mha_dec" value="Multi-Head&#xa;Attention" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#fff2cc;strokeColor=#d6b656;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="650" y="240" width="120" height="40" as="geometry"/>
</mxCell>
<mxCell id="add_norm2_dec" value="Add &amp; Norm" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#d5e8d4;strokeColor=#82b366;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="650" y="200" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="ff_dec" value="Feed Forward" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#fff2cc;strokeColor=#d6b656;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="650" y="160" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="add_norm3_dec" value="Add &amp; Norm" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#d5e8d4;strokeColor=#82b366;fontSize=10;" vertex="1" parent="1">
<mxGeometry x="650" y="120" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="nx_dec" value="Nx" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;fontStyle=2;" vertex="1" parent="1">
<mxGeometry x="790" y="255" width="30" height="30" as="geometry"/>
</mxCell>
<mxCell id="linear" value="Linear" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#f8cecc;strokeColor=#b85450;fontSize=11;" vertex="1" parent="1">
<mxGeometry x="650" y="80" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="softmax" value="Softmax" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#f8cecc;strokeColor=#b85450;fontSize=11;" vertex="1" parent="1">
<mxGeometry x="650" y="40" width="120" height="30" as="geometry"/>
</mxCell>
<mxCell id="output" value="Output Probabilities" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#dae8fc;strokeColor=#6c8ebf;fontSize=11;fontStyle=1;" vertex="1" parent="1">
<mxGeometry x="640" y="0" width="140" height="30" as="geometry"/>
</mxCell>
<mxCell id="conn1" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#6c8ebf;flowAnimation=1;" edge="1" parent="1" source="input_embed" target="pos_enc_left">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn2" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#6c8ebf;flowAnimation=1;" edge="1" parent="1" source="pos_enc_left" target="mha_enc">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn3" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#82b366;flowAnimation=1;" edge="1" parent="1" source="mha_enc" target="add_norm1_enc">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn4" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#d6b656;flowAnimation=1;" edge="1" parent="1" source="add_norm1_enc" target="ff_enc">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn5" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#82b366;flowAnimation=1;" edge="1" parent="1" source="ff_enc" target="add_norm2_enc">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn_cross" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;strokeWidth=3;strokeColor=#9673a6;flowAnimation=1;dashed=1;" edge="1" parent="1" source="add_norm2_enc" target="mha_dec">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="400" y="215"/>
<mxPoint x="400" y="260"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="cross_label" value="K, V" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];fontSize=10;fontStyle=1;fillColor=#ffffff;" vertex="1" connectable="0" parent="conn_cross">
<mxGeometry x="-0.1" y="1" relative="1" as="geometry">
<mxPoint x="10" y="-9" as="offset"/>
</mxGeometry>
</mxCell>
<mxCell id="conn6" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#d79b00;flowAnimation=1;" edge="1" parent="1" source="output_embed" target="pos_enc_right">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn7" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#d79b00;flowAnimation=1;" edge="1" parent="1" source="pos_enc_right" target="masked_mha_dec">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn8" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#82b366;flowAnimation=1;" edge="1" parent="1" source="masked_mha_dec" target="add_norm1_dec">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn9" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#d6b656;flowAnimation=1;" edge="1" parent="1" source="add_norm1_dec" target="mha_dec">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#82b366;flowAnimation=1;" edge="1" parent="1" source="mha_dec" target="add_norm2_dec">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#d6b656;flowAnimation=1;" edge="1" parent="1" source="add_norm2_dec" target="ff_dec">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#82b366;flowAnimation=1;" edge="1" parent="1" source="ff_dec" target="add_norm3_dec">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#b85450;flowAnimation=1;" edge="1" parent="1" source="add_norm3_dec" target="linear">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn14" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#b85450;flowAnimation=1;" edge="1" parent="1" source="linear" target="softmax">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="conn15" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;strokeWidth=2;strokeColor=#6c8ebf;flowAnimation=1;" edge="1" parent="1" source="softmax" target="output">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="res1_enc" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;strokeWidth=1.5;strokeColor=#999999;dashed=1;flowAnimation=1;" edge="1" parent="1" source="mha_enc" target="add_norm1_enc">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="50" y="355"/>
<mxPoint x="50" y="295"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="res2_enc" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;strokeWidth=1.5;strokeColor=#999999;dashed=1;flowAnimation=1;" edge="1" parent="1" source="ff_enc" target="add_norm2_enc">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="50" y="255"/>
<mxPoint x="50" y="215"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="res1_dec" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=1;entryY=0.5;entryDx=0;entryDy=0;strokeWidth=1.5;strokeColor=#999999;dashed=1;flowAnimation=1;" edge="1" parent="1" source="masked_mha_dec" target="add_norm1_dec">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="800" y="365"/>
<mxPoint x="800" y="305"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="res2_dec" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=1;entryY=0.5;entryDx=0;entryDy=0;strokeWidth=1.5;strokeColor=#999999;dashed=1;flowAnimation=1;" edge="1" parent="1" source="mha_dec" target="add_norm2_dec">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="800" y="260"/>
<mxPoint x="800" y="215"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="res3_dec" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=1;entryY=0.5;entryDx=0;entryDy=0;strokeWidth=1.5;strokeColor=#999999;dashed=1;flowAnimation=1;" edge="1" parent="1" source="ff_dec" target="add_norm3_dec">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="800" y="175"/>
<mxPoint x="800" y="135"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="input_label" value="Inputs" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=12;fontStyle=1;" vertex="1" parent="1">
<mxGeometry x="110" y="530" width="60" height="20" as="geometry"/>
</mxCell>
<mxCell id="output_label" value="Outputs&#xa;(shifted right)" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=12;fontStyle=1;" vertex="1" parent="1">
<mxGeometry x="660" y="530" width="100" height="30" as="geometry"/>
</mxCell>`,
},
{
promptText: "Replicate this in aws style",
hasImage: true,
xml: `<mxCell id="2" value="AWS" style="sketch=0;outlineConnect=0;gradientColor=none;html=1;whiteSpace=wrap;fontSize=12;fontStyle=0;container=1;pointerEvents=0;collapsible=0;recursiveResize=0;shape=mxgraph.aws4.group;grIcon=mxgraph.aws4.group_aws_cloud;strokeColor=#232F3E;fillColor=none;verticalAlign=top;align=left;spacingLeft=30;fontColor=#232F3E;dashed=0;rounded=1;arcSize=5;" vertex="1" parent="1">
<mxGeometry x="340" y="40" width="880" height="520" as="geometry"/>
</mxCell>
<mxCell id="3" value="User" style="sketch=0;outlineConnect=0;fontColor=#232F3E;gradientColor=none;fillColor=#232F3D;strokeColor=none;dashed=0;verticalLabelPosition=bottom;verticalAlign=top;align=center;html=1;fontSize=14;fontStyle=0;aspect=fixed;pointerEvents=1;shape=mxgraph.aws4.user;rounded=1;" vertex="1" parent="1">
<mxGeometry x="80" y="240" width="78" height="78" as="geometry"/>
</mxCell>
<mxCell id="4" value="EC2" style="sketch=0;points=[[0,0,0],[0.25,0,0],[0.5,0,0],[0.75,0,0],[1,0,0],[0,1,0],[0.25,1,0],[0.5,1,0],[0.75,1,0],[1,1,0],[0,0.25,0],[0,0.5,0],[0,0.75,0],[1,0.25,0],[1,0.5,0],[1,0.75,0]];outlineConnect=0;fontColor=#232F3E;fillColor=#ED7100;strokeColor=#ffffff;dashed=0;verticalLabelPosition=bottom;verticalAlign=top;align=center;html=1;fontSize=14;fontStyle=0;aspect=fixed;shape=mxgraph.aws4.resourceIcon;resIcon=mxgraph.aws4.ec2;rounded=1;" vertex="1" parent="1">
<mxGeometry x="560" y="240" width="78" height="78" as="geometry"/>
</mxCell>
<mxCell id="5" value="S3" style="sketch=0;points=[[0,0,0],[0.25,0,0],[0.5,0,0],[0.75,0,0],[1,0,0],[0,1,0],[0.25,1,0],[0.5,1,0],[0.75,1,0],[1,1,0],[0,0.25,0],[0,0.5,0],[0,0.75,0],[1,0.25,0],[1,0.5,0],[1,0.75,0]];outlineConnect=0;fontColor=#232F3E;fillColor=#7AA116;strokeColor=#ffffff;dashed=0;verticalLabelPosition=bottom;verticalAlign=top;align=center;html=1;fontSize=14;fontStyle=0;aspect=fixed;shape=mxgraph.aws4.resourceIcon;resIcon=mxgraph.aws4.s3;rounded=1;" vertex="1" parent="1">
<mxGeometry x="960" y="120" width="78" height="78" as="geometry"/>
</mxCell>
<mxCell id="6" value="bedrock" style="sketch=0;points=[[0,0,0],[0.25,0,0],[0.5,0,0],[0.75,0,0],[1,0,0],[0,1,0],[0.25,1,0],[0.5,1,0],[0.75,1,0],[1,1,0],[0,0.25,0],[0,0.5,0],[0,0.75,0],[1,0.25,0],[1,0.5,0],[1,0.75,0]];outlineConnect=0;fontColor=#232F3E;fillColor=#01A88D;strokeColor=#ffffff;dashed=0;verticalLabelPosition=bottom;verticalAlign=top;align=center;html=1;fontSize=14;fontStyle=0;aspect=fixed;shape=mxgraph.aws4.resourceIcon;resIcon=mxgraph.aws4.bedrock;rounded=1;" vertex="1" parent="1">
<mxGeometry x="960" y="260" width="78" height="78" as="geometry"/>
</mxCell>
<mxCell id="7" value="DynamoDB" style="sketch=0;points=[[0,0,0],[0.25,0,0],[0.5,0,0],[0.75,0,0],[1,0,0],[0,1,0],[0.25,1,0],[0.5,1,0],[0.75,1,0],[1,1,0],[0,0.25,0],[0,0.5,0],[0,0.75,0],[1,0.25,0],[1,0.5,0],[1,0.75,0]];outlineConnect=0;fontColor=#232F3E;fillColor=#C925D1;strokeColor=#ffffff;dashed=0;verticalLabelPosition=bottom;verticalAlign=top;align=center;html=1;fontSize=14;fontStyle=0;aspect=fixed;shape=mxgraph.aws4.resourceIcon;resIcon=mxgraph.aws4.dynamodb;rounded=1;" vertex="1" parent="1">
<mxGeometry x="960" y="400" width="78" height="78" as="geometry"/>
</mxCell>
<mxCell id="8" value="" style="endArrow=classic;html=1;rounded=0;strokeColor=#232F3E;strokeWidth=2;exitX=1;exitY=0.5;exitDx=0;exitDy=0;exitPerimeter=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;entryPerimeter=0;" edge="1" parent="1" source="3" target="4">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="400" y="350" as="sourcePoint"/>
<mxPoint x="450" y="300" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="9" value="" style="endArrow=classic;html=1;rounded=0;strokeColor=#232F3E;strokeWidth=2;exitX=1;exitY=0.25;exitDx=0;exitDy=0;exitPerimeter=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;entryPerimeter=0;" edge="1" parent="1" source="4" target="5">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="700" y="350" as="sourcePoint"/>
<mxPoint x="750" y="300" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="10" value="" style="endArrow=classic;html=1;rounded=0;strokeColor=#232F3E;strokeWidth=2;exitX=1;exitY=0.5;exitDx=0;exitDy=0;exitPerimeter=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;entryPerimeter=0;" edge="1" parent="1" source="4" target="6">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="700" y="350" as="sourcePoint"/>
<mxPoint x="750" y="300" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="11" value="" style="endArrow=classic;html=1;rounded=0;strokeColor=#232F3E;strokeWidth=2;exitX=1;exitY=0.75;exitDx=0;exitDy=0;exitPerimeter=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;entryPerimeter=0;" edge="1" parent="1" source="4" target="7">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="700" y="350" as="sourcePoint"/>
<mxPoint x="750" y="300" as="targetPoint"/>
</mxGeometry>
</mxCell>`,
},
{
promptText: "Replicate this flowchart.",
hasImage: true,
xml: `<mxCell id="2" value="Lamp doesn't work" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffcccc;strokeColor=#000000;strokeWidth=2;fontSize=18;fontStyle=0;" vertex="1" parent="1">
<mxGeometry x="140" y="40" width="180" height="60" as="geometry"/>
</mxCell>
<mxCell id="3" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;strokeColor=#000000;strokeWidth=2;endArrow=block;endFill=1;" edge="1" parent="1" source="2" target="4">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="4" value="Lamp&lt;br&gt;plugged in?" style="rhombus;whiteSpace=wrap;html=1;fillColor=#ffff99;strokeColor=#000000;strokeWidth=2;fontSize=18;fontStyle=0;" vertex="1" parent="1">
<mxGeometry x="130" y="150" width="200" height="200" as="geometry"/>
</mxCell>
<mxCell id="5" value="No" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;strokeColor=#000000;strokeWidth=2;endArrow=block;endFill=1;fontSize=16;" edge="1" parent="1" source="4" target="6">
<mxGeometry x="-0.2" relative="1" as="geometry">
<mxPoint as="offset"/>
</mxGeometry>
</mxCell>
<mxCell id="6" value="Plug in lamp" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#99ff99;strokeColor=#000000;strokeWidth=2;fontSize=18;fontStyle=0;" vertex="1" parent="1">
<mxGeometry x="420" y="220" width="200" height="60" as="geometry"/>
</mxCell>
<mxCell id="7" value="Yes" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;strokeColor=#000000;strokeWidth=2;endArrow=block;endFill=1;fontSize=16;" edge="1" parent="1" source="4" target="8">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="8" value="Bulb&lt;br&gt;burned out?" style="rhombus;whiteSpace=wrap;html=1;fillColor=#ffff99;strokeColor=#000000;strokeWidth=2;fontSize=18;fontStyle=0;" vertex="1" parent="1">
<mxGeometry x="130" y="400" width="200" height="200" as="geometry"/>
</mxCell>
<mxCell id="9" value="Yes" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;strokeColor=#000000;strokeWidth=2;endArrow=block;endFill=1;fontSize=16;" edge="1" parent="1" source="8" target="10">
<mxGeometry x="-0.2" relative="1" as="geometry">
<mxPoint as="offset"/>
</mxGeometry>
</mxCell>
<mxCell id="10" value="Replace bulb" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#99ff99;strokeColor=#000000;strokeWidth=2;fontSize=18;fontStyle=0;" vertex="1" parent="1">
<mxGeometry x="420" y="470" width="200" height="60" as="geometry"/>
</mxCell>
<mxCell id="11" value="No" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;strokeColor=#000000;strokeWidth=2;endArrow=block;endFill=1;fontSize=16;" edge="1" parent="1" source="8" target="12">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="12" value="Repair lamp" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#99ff99;strokeColor=#000000;strokeWidth=2;fontSize=18;fontStyle=0;" vertex="1" parent="1">
<mxGeometry x="130" y="650" width="200" height="60" as="geometry"/>
</mxCell>`,
},
{
promptText: "Summarize this paper as a diagram",
hasImage: true,
xml: `<mxCell id="title_bg" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#1a237e;strokeColor=none;arcSize=8;"
value="" vertex="1">
<mxGeometry height="80" width="720" x="40" y="20" as="geometry" />
</mxCell>
<mxCell id="title" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=22;fontStyle=1;fontColor=#FFFFFF;"
value="Chain-of-Thought Prompting&lt;br&gt;&lt;font style=&quot;font-size: 14px;&quot;&gt;Elicits Reasoning in Large Language Models&lt;/font&gt;"
vertex="1">
<mxGeometry height="70" width="720" x="40" y="25" as="geometry" />
</mxCell>
<mxCell id="authors" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;fontColor=#666666;"
value="Wei et al. (Google Research, Brain Team) | NeurIPS 2022" vertex="1">
<mxGeometry height="20" width="720" x="40" y="100" as="geometry" />
</mxCell>
<mxCell id="core_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="💡 Core Idea" vertex="1">
<mxGeometry height="30" width="150" x="40" y="125" as="geometry" />
</mxCell>
<mxCell id="core_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E3F2FD;strokeColor=#1565C0;align=left;spacingLeft=10;spacingRight=10;fontSize=11;"
value="&lt;b&gt;Chain of Thought&lt;/b&gt; = A series of intermediate reasoning steps that lead to the final answer&lt;br&gt;&lt;br&gt;Simply provide a few CoT demonstrations as exemplars in few-shot prompting"
vertex="1">
<mxGeometry height="75" width="340" x="40" y="155" as="geometry" />
</mxCell>
<mxCell id="compare_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="⚖️ Standard vs Chain-of-Thought Prompting" vertex="1">
<mxGeometry height="30" width="350" x="40" y="240" as="geometry" />
</mxCell>
<mxCell id="std_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#FFEBEE;strokeColor=#C62828;arcSize=8;"
value="" vertex="1">
<mxGeometry height="160" width="170" x="40" y="275" as="geometry" />
</mxCell>
<mxCell id="std_title" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=12;fontStyle=1;fontColor=#C62828;"
value="Standard Prompting" vertex="1">
<mxGeometry height="25" width="170" x="40" y="280" as="geometry" />
</mxCell>
<mxCell id="std_q" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=top;whiteSpace=wrap;rounded=0;fontSize=9;spacingLeft=5;spacingRight=5;"
value="Q: Roger has 5 tennis balls. He buys 2 more cans. Each can has 3 balls. How many now?"
vertex="1">
<mxGeometry height="55" width="160" x="45" y="305" as="geometry" />
</mxCell>
<mxCell id="std_a" parent="1"
style="text;html=1;strokeColor=none;fillColor=#FFCDD2;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=1;fontSize=10;fontStyle=1;spacingLeft=5;"
value="A: The answer is 11." vertex="1">
<mxGeometry height="25" width="150" x="50" y="365" as="geometry" />
</mxCell>
<mxCell id="std_result" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;fontStyle=1;fontColor=#C62828;"
value="❌ Often Wrong" vertex="1">
<mxGeometry height="30" width="170" x="40" y="400" as="geometry" />
</mxCell>
<mxCell id="cot_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E8F5E9;strokeColor=#2E7D32;arcSize=8;"
value="" vertex="1">
<mxGeometry height="160" width="170" x="220" y="275" as="geometry" />
</mxCell>
<mxCell id="cot_title" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=12;fontStyle=1;fontColor=#2E7D32;"
value="Chain-of-Thought" vertex="1">
<mxGeometry height="25" width="170" x="220" y="280" as="geometry" />
</mxCell>
<mxCell id="cot_q" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=top;whiteSpace=wrap;rounded=0;fontSize=9;spacingLeft=5;spacingRight=5;"
value="Q: Roger has 5 tennis balls. He buys 2 more cans. Each can has 3 balls. How many now?"
vertex="1">
<mxGeometry height="55" width="160" x="225" y="305" as="geometry" />
</mxCell>
<mxCell id="cot_a" parent="1"
style="text;html=1;strokeColor=none;fillColor=#C8E6C9;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=1;fontSize=9;fontStyle=1;spacingLeft=5;"
value="A: 2 cans × 3 = 6 balls.&lt;br&gt;5 + 6 = 11. Answer: 11" vertex="1">
<mxGeometry height="35" width="150" x="230" y="360" as="geometry" />
</mxCell>
<mxCell id="cot_result" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;fontStyle=1;fontColor=#2E7D32;"
value="✓ Correct!" vertex="1">
<mxGeometry height="30" width="170" x="220" y="400" as="geometry" />
</mxCell>
<mxCell id="vs_arrow" edge="1" parent="1"
style="shape=flexArrow;endArrow=classic;startArrow=classic;html=1;fillColor=#FFC107;strokeColor=none;width=8;endSize=4;startSize=4;"
value="">
<mxGeometry relative="1" width="100" as="geometry">
<mxPoint x="195" y="355" as="sourcePoint" />
<mxPoint x="235" y="355" as="targetPoint" />
</mxGeometry>
</mxCell>
<mxCell id="props_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="🔑 Key Properties" vertex="1">
<mxGeometry height="30" width="150" x="400" y="125" as="geometry" />
</mxCell>
<mxCell id="prop1" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#FFF3E0;strokeColor=#EF6C00;fontSize=10;align=left;spacingLeft=8;"
value="1⃣ Decomposes multi-step problems" vertex="1">
<mxGeometry height="32" width="180" x="400" y="155" as="geometry" />
</mxCell>
<mxCell id="prop2" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#FFF3E0;strokeColor=#EF6C00;fontSize=10;align=left;spacingLeft=8;"
value="2⃣ Interpretable reasoning window" vertex="1">
<mxGeometry height="32" width="180" x="400" y="192" as="geometry" />
</mxCell>
<mxCell id="prop3" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#FFF3E0;strokeColor=#EF6C00;fontSize=10;align=left;spacingLeft=8;"
value="3⃣ Applicable to any language task" vertex="1">
<mxGeometry height="32" width="180" x="400" y="229" as="geometry" />
</mxCell>
<mxCell id="prop4" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#FFF3E0;strokeColor=#EF6C00;fontSize=10;align=left;spacingLeft=8;"
value="4⃣ No finetuning required" vertex="1">
<mxGeometry height="32" width="180" x="400" y="266" as="geometry" />
</mxCell>
<mxCell id="emergent_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="📈 Emergent Ability" vertex="1">
<mxGeometry height="30" width="180" x="400" y="310" as="geometry" />
</mxCell>
<mxCell id="emergent_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#F3E5F5;strokeColor=#7B1FA2;arcSize=8;"
value="" vertex="1">
<mxGeometry height="95" width="180" x="400" y="340" as="geometry" />
</mxCell>
<mxCell id="emergent_text" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;"
value="CoT only works with&lt;br&gt;&lt;b&gt;~100B+ parameters&lt;/b&gt;&lt;br&gt;&lt;br&gt;Small models produce&lt;br&gt;fluent but illogical chains"
vertex="1">
<mxGeometry height="85" width="180" x="400" y="345" as="geometry" />
</mxCell>
<mxCell id="results_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="📊 Key Results" vertex="1">
<mxGeometry height="30" width="150" x="600" y="125" as="geometry" />
</mxCell>
<mxCell id="gsm_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E8F5E9;strokeColor=#2E7D32;arcSize=8;"
value="" vertex="1">
<mxGeometry height="100" width="160" x="600" y="155" as="geometry" />
</mxCell>
<mxCell id="gsm_title" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=12;fontStyle=1;fontColor=#2E7D32;"
value="GSM8K (Math)" vertex="1">
<mxGeometry height="20" width="160" x="600" y="160" as="geometry" />
</mxCell>
<mxCell id="gsm_bar1" parent="1"
style="rounded=0;whiteSpace=wrap;html=1;fillColor=#FFCDD2;strokeColor=none;"
value="" vertex="1">
<mxGeometry height="30" width="40" x="615" y="185" as="geometry" />
</mxCell>
<mxCell id="gsm_bar2" parent="1"
style="rounded=0;whiteSpace=wrap;html=1;fillColor=#4CAF50;strokeColor=none;"
value="" vertex="1">
<mxGeometry height="30" width="80" x="665" y="185" as="geometry" />
</mxCell>
<mxCell id="gsm_label1" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;fontStyle=1;"
value="18%" vertex="1">
<mxGeometry height="15" width="40" x="615" y="215" as="geometry" />
</mxCell>
<mxCell id="gsm_label2" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;fontStyle=1;fontColor=#2E7D32;"
value="57%" vertex="1">
<mxGeometry height="15" width="80" x="665" y="215" as="geometry" />
</mxCell>
<mxCell id="gsm_legend" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=9;fontColor=#666666;"
value="Standard → CoT (PaLM 540B)" vertex="1">
<mxGeometry height="20" width="160" x="600" y="232" as="geometry" />
</mxCell>
<mxCell id="bench_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="🧪 Benchmarks Tested" vertex="1">
<mxGeometry height="30" width="180" x="600" y="265" as="geometry" />
</mxCell>
<mxCell id="bench_arith" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E3F2FD;strokeColor=#1565C0;fontSize=10;align=center;"
value="🔢 Arithmetic&lt;br&gt;&lt;font style=&quot;font-size: 9px;&quot;&gt;GSM8K, SVAMP, ASDiv, AQuA, MAWPS&lt;/font&gt;"
vertex="1">
<mxGeometry height="45" width="160" x="600" y="295" as="geometry" />
</mxCell>
<mxCell id="bench_common" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E3F2FD;strokeColor=#1565C0;fontSize=10;align=center;"
value="🧠 Commonsense&lt;br&gt;&lt;font style=&quot;font-size: 9px;&quot;&gt;CSQA, StrategyQA, Date, Sports, SayCan&lt;/font&gt;"
vertex="1">
<mxGeometry height="45" width="160" x="600" y="345" as="geometry" />
</mxCell>
<mxCell id="bench_symbol" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E3F2FD;strokeColor=#1565C0;fontSize=10;align=center;"
value="🔣 Symbolic&lt;br&gt;&lt;font style=&quot;font-size: 9px;&quot;&gt;Last Letter Concat, Coin Flip&lt;/font&gt;"
vertex="1">
<mxGeometry height="40" width="160" x="600" y="395" as="geometry" />
</mxCell>
<mxCell id="task_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="🎯 Task Types &amp; Results" vertex="1">
<mxGeometry height="30" width="200" x="40" y="445" as="geometry" />
</mxCell>
<mxCell id="task_arith" parent="1"
style="ellipse;whiteSpace=wrap;html=1;fillColor=#BBDEFB;strokeColor=#1565C0;fontSize=11;fontStyle=1;"
value="Arithmetic&lt;br&gt;Reasoning" vertex="1">
<mxGeometry height="60" width="90" x="40" y="480" as="geometry" />
</mxCell>
<mxCell id="task_arith_res" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=top;whiteSpace=wrap;rounded=0;fontSize=9;fontColor=#1565C0;"
value="SOTA on GSM8K&lt;br&gt;(57% vs 55% prior)" vertex="1">
<mxGeometry height="30" width="110" x="30" y="540" as="geometry" />
</mxCell>
<mxCell id="task_common" parent="1"
style="ellipse;whiteSpace=wrap;html=1;fillColor=#C8E6C9;strokeColor=#2E7D32;fontSize=11;fontStyle=1;"
value="Commonsense&lt;br&gt;Reasoning" vertex="1">
<mxGeometry height="60" width="90" x="160" y="480" as="geometry" />
</mxCell>
<mxCell id="task_common_res" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=top;whiteSpace=wrap;rounded=0;fontSize=9;fontColor=#2E7D32;"
value="SOTA StrategyQA&lt;br&gt;(75.6% vs 69.4%)" vertex="1">
<mxGeometry height="30" width="110" x="150" y="540" as="geometry" />
</mxCell>
<mxCell id="task_symbol" parent="1"
style="ellipse;whiteSpace=wrap;html=1;fillColor=#FFE0B2;strokeColor=#EF6C00;fontSize=11;fontStyle=1;"
value="Symbolic&lt;br&gt;Reasoning" vertex="1">
<mxGeometry height="60" width="90" x="280" y="480" as="geometry" />
</mxCell>
<mxCell id="task_symbol_res" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=top;whiteSpace=wrap;rounded=0;fontSize=9;fontColor=#EF6C00;"
value="OOD Generalization&lt;br&gt;to longer sequences" vertex="1">
<mxGeometry height="30" width="110" x="270" y="540" as="geometry" />
</mxCell>
<mxCell id="task_arrow1" edge="1" parent="1"
style="endArrow=classic;html=1;strokeColor=#9E9E9E;strokeWidth=2;" value="">
<mxGeometry height="50" relative="1" width="50" as="geometry">
<mxPoint x="130" y="510" as="sourcePoint" />
<mxPoint x="160" y="510" as="targetPoint" />
</mxGeometry>
</mxCell>
<mxCell id="task_arrow2" edge="1" parent="1"
style="endArrow=classic;html=1;strokeColor=#9E9E9E;strokeWidth=2;" value="">
<mxGeometry height="50" relative="1" width="50" as="geometry">
<mxPoint x="250" y="510" as="sourcePoint" />
<mxPoint x="280" y="510" as="targetPoint" />
</mxGeometry>
</mxCell>
<mxCell id="models_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="🤖 Models Tested" vertex="1">
<mxGeometry height="30" width="150" x="400" y="445" as="geometry" />
</mxCell>
<mxCell id="models_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ECEFF1;strokeColor=#607D8B;arcSize=8;"
value="" vertex="1">
<mxGeometry height="95" width="180" x="400" y="475" as="geometry" />
</mxCell>
<mxCell id="model1" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;spacingLeft=10;"
value="• GPT-3 (175B)" vertex="1">
<mxGeometry height="20" width="90" x="400" y="480" as="geometry" />
</mxCell>
<mxCell id="model2" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;spacingLeft=10;"
value="• LaMDA (137B)" vertex="1">
<mxGeometry height="20" width="90" x="400" y="500" as="geometry" />
</mxCell>
<mxCell id="model3" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;spacingLeft=10;"
value="• PaLM (540B)" vertex="1">
<mxGeometry height="20" width="90" x="400" y="520" as="geometry" />
</mxCell>
<mxCell id="model4" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;spacingLeft=10;"
value="• Codex" vertex="1">
<mxGeometry height="20" width="80" x="490" y="480" as="geometry" />
</mxCell>
<mxCell id="model5" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=11;spacingLeft=10;"
value="• UL2 (20B)" vertex="1">
<mxGeometry height="20" width="80" x="490" y="500" as="geometry" />
</mxCell>
<mxCell id="model_note" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;fontStyle=2;fontColor=#607D8B;"
value="No finetuning - prompting only!" vertex="1">
<mxGeometry height="20" width="180" x="400" y="545" as="geometry" />
</mxCell>
<mxCell id="takeaway_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=16;fontStyle=1;fontColor=#1a237e;"
value="✨ Key Takeaways" vertex="1">
<mxGeometry height="30" width="160" x="600" y="445" as="geometry" />
</mxCell>
<mxCell id="takeaway_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#FFF8E1;strokeColor=#FFA000;arcSize=8;"
value="" vertex="1">
<mxGeometry height="95" width="160" x="600" y="475" as="geometry" />
</mxCell>
<mxCell id="take1" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;spacingLeft=5;"
value="✓ Simple yet powerful" vertex="1">
<mxGeometry height="18" width="150" x="605" y="480" as="geometry" />
</mxCell>
<mxCell id="take2" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;spacingLeft=5;"
value="✓ Emergent at scale" vertex="1">
<mxGeometry height="18" width="150" x="605" y="498" as="geometry" />
</mxCell>
<mxCell id="take3" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;spacingLeft=5;"
value="✓ Broadly applicable" vertex="1">
<mxGeometry height="18" width="150" x="605" y="516" as="geometry" />
</mxCell>
<mxCell id="take4" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;spacingLeft=5;"
value="✓ No training needed" vertex="1">
<mxGeometry height="18" width="150" x="605" y="534" as="geometry" />
</mxCell>
<mxCell id="take5" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=10;spacingLeft=5;"
value="✓ State-of-the-art results" vertex="1">
<mxGeometry height="18" width="150" x="605" y="552" as="geometry" />
</mxCell>
<mxCell id="format_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=14;fontStyle=1;fontColor=#1a237e;"
value="📝 Prompt Format" vertex="1">
<mxGeometry height="25" width="150" x="40" y="575" as="geometry" />
</mxCell>
<mxCell id="format_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E1BEE7;strokeColor=#7B1FA2;fontSize=12;fontStyle=1;"
value="〈 Input, Chain of Thought, Output 〉" vertex="1">
<mxGeometry height="35" width="250" x="40" y="600" as="geometry" />
</mxCell>
<mxCell id="limit_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=14;fontStyle=1;fontColor=#1a237e;"
value="⚠️ Limitations" vertex="1">
<mxGeometry height="25" width="120" x="310" y="575" as="geometry" />
</mxCell>
<mxCell id="limit_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#FFEBEE;strokeColor=#C62828;fontSize=10;align=left;spacingLeft=8;"
value="• Requires large models (~100B+)&lt;br&gt;• No guarantee of correct reasoning&lt;br&gt;• Costly to serve in production"
vertex="1">
<mxGeometry height="55" width="200" x="310" y="600" as="geometry" />
</mxCell>
<mxCell id="impact_header" parent="1"
style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontSize=14;fontStyle=1;fontColor=#1a237e;"
value="🚀 Impact" vertex="1">
<mxGeometry height="25" width="100" x="530" y="575" as="geometry" />
</mxCell>
<mxCell id="impact_box" parent="1"
style="rounded=1;whiteSpace=wrap;html=1;fillColor=#E8F5E9;strokeColor=#2E7D32;fontSize=10;align=left;spacingLeft=8;spacingRight=8;"
value="Foundational technique for modern LLM reasoning - inspired many follow-up works including Self-Consistency, Tree-of-Thought, etc."
vertex="1">
<mxGeometry height="55" width="230" x="530" y="600" as="geometry" />
</mxCell>`,
},
{
promptText: "Draw a cat for me",
hasImage: false,
xml: `<mxCell id="2" value="" style="ellipse;whiteSpace=wrap;html=1;aspect=fixed;fillColor=#FFE6CC;strokeColor=#000000;strokeWidth=2;" vertex="1" parent="1">
<mxGeometry x="300" y="150" width="120" height="120" as="geometry"/>
</mxCell>
<mxCell id="3" value="" style="triangle;whiteSpace=wrap;html=1;fillColor=#FFE6CC;strokeColor=#000000;strokeWidth=2;rotation=30;" vertex="1" parent="1">
<mxGeometry x="280" y="120" width="50" height="60" as="geometry"/>
</mxCell>
<mxCell id="4" value="" style="triangle;whiteSpace=wrap;html=1;fillColor=#FFE6CC;strokeColor=#000000;strokeWidth=2;rotation=-30;" vertex="1" parent="1">
<mxGeometry x="390" y="120" width="50" height="60" as="geometry"/>
</mxCell>
<mxCell id="5" value="" style="triangle;whiteSpace=wrap;html=1;fillColor=#FFB6C1;strokeColor=none;rotation=30;" vertex="1" parent="1">
<mxGeometry x="290" y="135" width="30" height="35" as="geometry"/>
</mxCell>
<mxCell id="6" value="" style="triangle;whiteSpace=wrap;html=1;fillColor=#FFB6C1;strokeColor=none;rotation=-30;" vertex="1" parent="1">
<mxGeometry x="400" y="135" width="30" height="35" as="geometry"/>
</mxCell>
<mxCell id="7" value="" style="ellipse;whiteSpace=wrap;html=1;aspect=fixed;fillColor=#000000;strokeColor=#000000;" vertex="1" parent="1">
<mxGeometry x="325" y="185" width="15" height="15" as="geometry"/>
</mxCell>
<mxCell id="8" value="" style="ellipse;whiteSpace=wrap;html=1;aspect=fixed;fillColor=#000000;strokeColor=#000000;" vertex="1" parent="1">
<mxGeometry x="380" y="185" width="15" height="15" as="geometry"/>
</mxCell>
<mxCell id="9" value="" style="triangle;whiteSpace=wrap;html=1;fillColor=#FFB6C1;strokeColor=#000000;strokeWidth=1;rotation=180;" vertex="1" parent="1">
<mxGeometry x="350" y="210" width="20" height="15" as="geometry"/>
</mxCell>
<mxCell id="10" value="" style="curved=1;endArrow=none;html=1;strokeColor=#000000;strokeWidth=2;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="360" y="220" as="sourcePoint"/>
<mxPoint x="340" y="235" as="targetPoint"/>
<Array as="points">
<mxPoint x="355" y="230"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="11" value="" style="curved=1;endArrow=none;html=1;strokeColor=#000000;strokeWidth=2;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="360" y="220" as="sourcePoint"/>
<mxPoint x="380" y="235" as="targetPoint"/>
<Array as="points">
<mxPoint x="365" y="230"/>
</Array>
</mxGeometry>
</mxCell>
<mxCell id="12" value="" style="endArrow=none;html=1;strokeColor=#000000;strokeWidth=1.5;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="310" y="200" as="sourcePoint"/>
<mxPoint x="260" y="195" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="13" value="" style="endArrow=none;html=1;strokeColor=#000000;strokeWidth=1.5;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="310" y="210" as="sourcePoint"/>
<mxPoint x="260" y="210" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="14" value="" style="endArrow=none;html=1;strokeColor=#000000;strokeWidth=1.5;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="310" y="220" as="sourcePoint"/>
<mxPoint x="260" y="225" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="15" value="" style="endArrow=none;html=1;strokeColor=#000000;strokeWidth=1.5;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="410" y="200" as="sourcePoint"/>
<mxPoint x="460" y="195" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="16" value="" style="endArrow=none;html=1;strokeColor=#000000;strokeWidth=1.5;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="410" y="210" as="sourcePoint"/>
<mxPoint x="460" y="210" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="17" value="" style="endArrow=none;html=1;strokeColor=#000000;strokeWidth=1.5;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="410" y="220" as="sourcePoint"/>
<mxPoint x="460" y="225" as="targetPoint"/>
</mxGeometry>
</mxCell>
<mxCell id="18" value="" style="ellipse;whiteSpace=wrap;html=1;fillColor=#FFE6CC;strokeColor=#000000;strokeWidth=2;" vertex="1" parent="1">
<mxGeometry x="285" y="250" width="150" height="180" as="geometry"/>
</mxCell>
<mxCell id="19" value="" style="ellipse;whiteSpace=wrap;html=1;fillColor=#FFFFFF;strokeColor=none;" vertex="1" parent="1">
<mxGeometry x="315" y="280" width="90" height="120" as="geometry"/>
</mxCell>
<mxCell id="20" value="" style="ellipse;whiteSpace=wrap;html=1;fillColor=#FFE6CC;strokeColor=#000000;strokeWidth=2;" vertex="1" parent="1">
<mxGeometry x="300" y="410" width="40" height="50" as="geometry"/>
</mxCell>
<mxCell id="21" value="" style="ellipse;whiteSpace=wrap;html=1;fillColor=#FFE6CC;strokeColor=#000000;strokeWidth=2;" vertex="1" parent="1">
<mxGeometry x="380" y="410" width="40" height="50" as="geometry"/>
</mxCell>
<mxCell id="22" value="" style="curved=1;endArrow=none;html=1;strokeColor=#000000;strokeWidth=3;fillColor=#FFE6CC;" edge="1" parent="1">
<mxGeometry width="50" height="50" relative="1" as="geometry">
<mxPoint x="285" y="340" as="sourcePoint"/>
<mxPoint x="240" y="260" as="targetPoint"/>
<Array as="points">
<mxPoint x="260" y="350"/>
<mxPoint x="240" y="320"/>
<mxPoint x="235" y="290"/>
</Array>
</mxGeometry>
</mxCell>`,
},
]
export function findCachedResponse(
promptText: string,
hasImage: boolean,
): CachedResponse | undefined {
return CACHED_EXAMPLE_RESPONSES.find(
(c) =>
c.promptText === promptText &&
c.hasImage === hasImage &&
c.xml !== "",
)
}

105
lib/langfuse.ts Normal file
View File

@@ -0,0 +1,105 @@
import { LangfuseClient } from "@langfuse/client"
import { observe, updateActiveTrace } from "@langfuse/tracing"
import * as api from "@opentelemetry/api"
// Singleton LangfuseClient instance for direct API calls
let langfuseClient: LangfuseClient | null = null
export function getLangfuseClient(): LangfuseClient | null {
if (!process.env.LANGFUSE_PUBLIC_KEY || !process.env.LANGFUSE_SECRET_KEY) {
return null
}
if (!langfuseClient) {
langfuseClient = new LangfuseClient({
publicKey: process.env.LANGFUSE_PUBLIC_KEY,
secretKey: process.env.LANGFUSE_SECRET_KEY,
baseUrl: process.env.LANGFUSE_BASEURL,
})
}
return langfuseClient
}
// Check if Langfuse is configured
export function isLangfuseEnabled(): boolean {
return !!process.env.LANGFUSE_PUBLIC_KEY
}
// Update trace with input data at the start of request
export function setTraceInput(params: {
input: string
sessionId?: string
userId?: string
}) {
if (!isLangfuseEnabled()) return
updateActiveTrace({
name: "chat",
input: params.input,
sessionId: params.sessionId,
userId: params.userId,
})
}
// Update trace with output and end the span
export function setTraceOutput(
output: string,
usage?: { promptTokens?: number; completionTokens?: number },
) {
if (!isLangfuseEnabled()) return
updateActiveTrace({ output })
const activeSpan = api.trace.getActiveSpan()
if (activeSpan) {
// Manually set usage attributes since AI SDK Bedrock streaming doesn't provide them
if (usage?.promptTokens) {
activeSpan.setAttribute("ai.usage.promptTokens", usage.promptTokens)
activeSpan.setAttribute(
"gen_ai.usage.input_tokens",
usage.promptTokens,
)
}
if (usage?.completionTokens) {
activeSpan.setAttribute(
"ai.usage.completionTokens",
usage.completionTokens,
)
activeSpan.setAttribute(
"gen_ai.usage.output_tokens",
usage.completionTokens,
)
}
activeSpan.end()
}
}
// Get telemetry config for streamText
export function getTelemetryConfig(params: {
sessionId?: string
userId?: string
}) {
if (!isLangfuseEnabled()) return undefined
return {
isEnabled: true,
recordInputs: true,
recordOutputs: true,
metadata: {
sessionId: params.sessionId,
userId: params.userId,
},
}
}
// Wrap a handler with Langfuse observe
export function wrapWithObserve<T>(
handler: (req: Request) => Promise<T>,
): (req: Request) => Promise<T> {
if (!isLangfuseEnabled()) {
return handler
}
return observe(handler, { name: "chat", endOnExit: false })
}

75
lib/pdf-utils.ts Normal file
View File

@@ -0,0 +1,75 @@
import { extractText, getDocumentProxy } from "unpdf"
// Maximum characters allowed for extracted text (configurable via env)
const DEFAULT_MAX_EXTRACTED_CHARS = 150000 // 150k chars
export const MAX_EXTRACTED_CHARS =
Number(process.env.NEXT_PUBLIC_MAX_EXTRACTED_CHARS) ||
DEFAULT_MAX_EXTRACTED_CHARS
// Text file extensions we support
const TEXT_EXTENSIONS = [
".txt",
".md",
".markdown",
".json",
".csv",
".xml",
".html",
".css",
".js",
".ts",
".jsx",
".tsx",
".py",
".java",
".c",
".cpp",
".h",
".go",
".rs",
".yaml",
".yml",
".toml",
".ini",
".log",
".sh",
".bash",
".zsh",
]
/**
* Extract text content from a PDF file
* Uses unpdf library for client-side extraction
*/
export async function extractPdfText(file: File): Promise<string> {
const buffer = await file.arrayBuffer()
const pdf = await getDocumentProxy(new Uint8Array(buffer))
const { text } = await extractText(pdf, { mergePages: true })
return text as string
}
/**
* Check if a file is a PDF
*/
export function isPdfFile(file: File): boolean {
return file.type === "application/pdf" || file.name.endsWith(".pdf")
}
/**
* Check if a file is a text file
*/
export function isTextFile(file: File): boolean {
const name = file.name.toLowerCase()
return (
file.type.startsWith("text/") ||
file.type === "application/json" ||
TEXT_EXTENSIONS.some((ext) => name.endsWith(ext))
)
}
/**
* Extract text content from a text file
*/
export async function extractTextFileContent(file: File): Promise<string> {
return await file.text()
}

27
lib/storage.ts Normal file
View File

@@ -0,0 +1,27 @@
// Centralized localStorage keys
// Consolidates all storage keys from chat-panel.tsx and settings-dialog.tsx
export const STORAGE_KEYS = {
// Chat data
messages: "next-ai-draw-io-messages",
xmlSnapshots: "next-ai-draw-io-xml-snapshots",
diagramXml: "next-ai-draw-io-diagram-xml",
sessionId: "next-ai-draw-io-session-id",
// Quota tracking
requestCount: "next-ai-draw-io-request-count",
requestDate: "next-ai-draw-io-request-date",
tokenCount: "next-ai-draw-io-token-count",
tokenDate: "next-ai-draw-io-token-date",
tpmCount: "next-ai-draw-io-tpm-count",
tpmMinute: "next-ai-draw-io-tpm-minute",
// Settings
accessCode: "next-ai-draw-io-access-code",
closeProtection: "next-ai-draw-io-close-protection",
accessCodeRequired: "next-ai-draw-io-access-code-required",
aiProvider: "next-ai-draw-io-ai-provider",
aiBaseUrl: "next-ai-draw-io-ai-base-url",
aiApiKey: "next-ai-draw-io-ai-api-key",
aiModel: "next-ai-draw-io-ai-model",
} as const

432
lib/system-prompts.ts Normal file
View File

@@ -0,0 +1,432 @@
/**
* System prompts for different AI models
* Extended prompt is used for models with higher cache token minimums (Opus 4.5, Haiku 4.5)
*
* Token counting utilities are in a separate file (token-counter.ts) to avoid
* WebAssembly issues with Next.js server-side rendering.
*/
// Default system prompt (~1900 tokens) - works with all models
export const DEFAULT_SYSTEM_PROMPT = `
You are an expert diagram creation assistant specializing in draw.io XML generation.
Your primary function is chat with user and crafting clear, well-organized visual diagrams through precise XML specifications.
You can see images that users upload, and you can read the text content extracted from PDF documents they upload.
When you are asked to create a diagram, briefly describe your plan about the layout and structure to avoid object overlapping or edge cross the objects. (2-3 sentences max), then use display_diagram tool to generate the XML.
After generating or editing a diagram, you don't need to say anything. The user can see the diagram - no need to describe it.
## App Context
You are an AI agent (powered by {{MODEL_NAME}}) inside a web app. The interface has:
- **Left panel**: Draw.io diagram editor where diagrams are rendered
- **Right panel**: Chat interface where you communicate with the user
You can read and modify diagrams by generating draw.io XML code through tool calls.
## App Features
1. **Diagram History** (clock icon, bottom-left of chat input): The app automatically saves a snapshot before each AI edit. Users can view the history panel and restore any previous version. Feel free to make changes - nothing is permanently lost.
2. **Theme Toggle** (palette icon, bottom-left of chat input): Users can switch between minimal UI and sketch-style UI for the draw.io editor.
3. **Image/PDF Upload** (paperclip icon, bottom-left of chat input): Users can upload images or PDF documents for you to analyze and generate diagrams from.
4. **Export** (via draw.io toolbar): Users can save diagrams as .drawio, .svg, or .png files.
5. **Clear Chat** (trash icon, bottom-right of chat input): Clears the conversation and resets the diagram.
You utilize the following tools:
---Tool1---
tool name: display_diagram
description: Display a NEW diagram on draw.io. Use this when creating a diagram from scratch or when major structural changes are needed.
parameters: {
xml: string
}
---Tool2---
tool name: edit_diagram
description: Edit specific parts of the EXISTING diagram. Use this when making small targeted changes like adding/removing elements, changing labels, or adjusting properties. This is more efficient than regenerating the entire diagram.
parameters: {
edits: Array<{search: string, replace: string}>
}
---Tool3---
tool name: append_diagram
description: Continue generating diagram XML when display_diagram was truncated due to output length limits. Only use this after display_diagram truncation.
parameters: {
xml: string // Continuation fragment (NO wrapper tags like <mxGraphModel> or <root>)
}
---End of tools---
IMPORTANT: Choose the right tool:
- Use display_diagram for: Creating new diagrams, major restructuring, or when the current diagram XML is empty
- Use edit_diagram for: Small modifications, adding/removing elements, changing text/colors, repositioning items
- Use append_diagram for: ONLY when display_diagram was truncated due to output length - continue generating from where you stopped
Core capabilities:
- Generate valid, well-formed XML strings for draw.io diagrams
- Create professional flowcharts, mind maps, entity diagrams, and technical illustrations
- Convert user descriptions into visually appealing diagrams using basic shapes and connectors
- Apply proper spacing, alignment and visual hierarchy in diagram layouts
- Adapt artistic concepts into abstract diagram representations using available shapes
- Optimize element positioning to prevent overlapping and maintain readability
- Structure complex systems into clear, organized visual components
Layout constraints:
- CRITICAL: Keep all diagram elements within a single page viewport to avoid page breaks
- Position all elements with x coordinates between 0-800 and y coordinates between 0-600
- Maximum width for containers (like AWS cloud boxes): 700 pixels
- Maximum height for containers: 550 pixels
- Use compact, efficient layouts that fit the entire diagram in one view
- Start positioning from reasonable margins (e.g., x=40, y=40) and keep elements grouped closely
- For large diagrams with many elements, use vertical stacking or grid layouts that stay within bounds
- Avoid spreading elements too far apart horizontally - users should see the complete diagram without a page break line
Note that:
- Use proper tool calls to generate or edit diagrams;
- never return raw XML in text responses,
- never use display_diagram to generate messages that you want to send user directly. e.g. to generate a "hello" text box when you want to greet user.
- Focus on producing clean, professional diagrams that effectively communicate the intended information through thoughtful layout and design choices.
- When artistic drawings are requested, creatively compose them using standard diagram shapes and connectors while maintaining visual clarity.
- Return XML only via tool calls, never in text responses.
- If user asks you to replicate a diagram based on an image, remember to match the diagram style and layout as closely as possible. Especially, pay attention to the lines and shapes, for example, if the lines are straight or curved, and if the shapes are rounded or square.
- Note that when you need to generate diagram about aws architecture, use **AWS 2025 icons**.
- NEVER include XML comments (<!-- ... -->) in your generated XML. Draw.io strips comments, which breaks edit_diagram patterns.
When using edit_diagram tool:
- CRITICAL: Copy search patterns EXACTLY from the "Current diagram XML" in system context - attribute order matters!
- Always include the element's id attribute for unique targeting: {"search": "<mxCell id=\\"5\\"", ...}
- Include complete elements (mxCell + mxGeometry) for reliable matching
- Preserve exact whitespace, indentation, and line breaks
- BAD: {"search": "value=\\"Label\\"", ...} - too vague, matches multiple elements
- GOOD: {"search": "<mxCell id=\\"3\\" value=\\"Old\\" style=\\"...\\">", "replace": "<mxCell id=\\"3\\" value=\\"New\\" style=\\"...\\">"}
- For multiple changes, use separate edits in array
- RETRY POLICY: If pattern not found, retry up to 3 times with adjusted patterns. After 3 failures, use display_diagram instead.
⚠️ CRITICAL JSON ESCAPING: When outputting edit_diagram tool calls, you MUST escape ALL double quotes inside string values:
- CORRECT: "y=\\"119\\"" (both quotes escaped)
- WRONG: "y="119\\"" (missing backslash before first quote - causes JSON parse error!)
- Every " inside a JSON string value needs \\" - no exceptions!
## Draw.io XML Structure Reference
**IMPORTANT:** You only generate the mxCell elements. The wrapper structure and root cells (id="0", id="1") are added automatically.
Example - generate ONLY this:
\`\`\`xml
<mxCell id="2" value="Label" style="rounded=1;" vertex="1" parent="1">
<mxGeometry x="100" y="100" width="120" height="60" as="geometry"/>
</mxCell>
\`\`\`
CRITICAL RULES:
1. Generate ONLY mxCell elements - NO wrapper tags (<mxfile>, <mxGraphModel>, <root>)
2. Do NOT include root cells (id="0" or id="1") - they are added automatically
3. ALL mxCell elements must be siblings - NEVER nest mxCell inside another mxCell
4. Use unique sequential IDs starting from "2"
5. Set parent="1" for top-level shapes, or parent="<container-id>" for grouped elements
Shape (vertex) example:
\`\`\`xml
<mxCell id="2" value="Label" style="rounded=1;whiteSpace=wrap;html=1;" vertex="1" parent="1">
<mxGeometry x="100" y="100" width="120" height="60" as="geometry"/>
</mxCell>
\`\`\`
Connector (edge) example:
\`\`\`xml
<mxCell id="3" style="endArrow=classic;html=1;" edge="1" parent="1" source="2" target="4">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
### Edge Routing Rules:
When creating edges/connectors, you MUST follow these rules to avoid overlapping lines:
**Rule 1: NEVER let multiple edges share the same path**
- If two edges connect the same pair of nodes, they MUST exit/enter at DIFFERENT positions
- Use exitY=0.3 for first edge, exitY=0.7 for second edge (NOT both 0.5)
**Rule 2: For bidirectional connections (A↔B), use OPPOSITE sides**
- A→B: exit from RIGHT side of A (exitX=1), enter LEFT side of B (entryX=0)
- B→A: exit from LEFT side of B (exitX=0), enter RIGHT side of A (entryX=1)
**Rule 3: Always specify exitX, exitY, entryX, entryY explicitly**
- Every edge MUST have these 4 attributes set in the style
- Example: style="edgeStyle=orthogonalEdgeStyle;exitX=1;exitY=0.3;entryX=0;entryY=0.3;endArrow=classic;"
**Rule 4: Route edges AROUND intermediate shapes (obstacle avoidance) - CRITICAL!**
- Before creating an edge, identify ALL shapes positioned between source and target
- If any shape is in the direct path, you MUST use waypoints to route around it
- For DIAGONAL connections: route along the PERIMETER (outside edge) of the diagram, NOT through the middle
- Add 20-30px clearance from shape boundaries when calculating waypoint positions
- Route ABOVE (lower y), BELOW (higher y), or to the SIDE of obstacles
- NEVER draw a line that visually crosses over another shape's bounding box
**Rule 5: Plan layout strategically BEFORE generating XML**
- Organize shapes into visual layers/zones (columns or rows) based on diagram flow
- Space shapes 150-200px apart to create clear routing channels for edges
- Mentally trace each edge: "What shapes are between source and target?"
- Prefer layouts where edges naturally flow in one direction (left-to-right or top-to-bottom)
**Rule 6: Use multiple waypoints for complex routing**
- One waypoint is often not enough - use 2-3 waypoints to create proper L-shaped or U-shaped paths
- Each direction change needs a waypoint (corner point)
- Waypoints should form clear horizontal/vertical segments (orthogonal routing)
- Calculate positions by: (1) identify obstacle boundaries, (2) add 20-30px margin
**Rule 7: Choose NATURAL connection points based on flow direction**
- NEVER use corner connections (e.g., entryX=1,entryY=1) - they look unnatural
- For TOP-TO-BOTTOM flow: exit from bottom (exitY=1), enter from top (entryY=0)
- For LEFT-TO-RIGHT flow: exit from right (exitX=1), enter from left (entryX=0)
- For DIAGONAL connections: use the side closest to the target, not corners
- Example: Node below-right of source → exit from bottom (exitY=1) OR right (exitX=1), not corner
**Before generating XML, mentally verify:**
1. "Do any edges cross over shapes that aren't their source/target?" → If yes, add waypoints
2. "Do any two edges share the same path?" → If yes, adjust exit/entry points
3. "Are any connection points at corners (both X and Y are 0 or 1)?" → If yes, use edge centers instead
4. "Could I rearrange shapes to reduce edge crossings?" → If yes, revise layout
\`\`\`
`
// Style instructions - only included when minimalStyle is false
const STYLE_INSTRUCTIONS = `
Common styles:
- Shapes: rounded=1 (rounded corners), fillColor=#hex, strokeColor=#hex
- Edges: endArrow=classic/block/open/none, startArrow=none/classic, curved=1, edgeStyle=orthogonalEdgeStyle
- Text: fontSize=14, fontStyle=1 (bold), align=center/left/right
`
// Minimal style instruction - skip styling and focus on layout (prepended to prompt for emphasis)
const MINIMAL_STYLE_INSTRUCTION = `
## ⚠️ MINIMAL STYLE MODE ACTIVE ⚠️
### No Styling - Plain Black/White Only
- NO fillColor, NO strokeColor, NO rounded, NO fontSize, NO fontStyle
- NO color attributes (no hex colors like #ff69b4)
- Style: "whiteSpace=wrap;html=1;" for shapes, "html=1;endArrow=classic;" for edges
- IGNORE all color/style examples below
### Container/Group Shapes - MUST be Transparent
- For container shapes (boxes that contain other shapes): use "fillColor=none;" to make background transparent
- This prevents containers from covering child elements
- Example: style="whiteSpace=wrap;html=1;fillColor=none;" for container rectangles
### Focus on Layout Quality
Since we skip styling, STRICTLY follow the "Edge Routing Rules" section below:
- SPACING: Minimum 50px gap between all elements
- NO OVERLAPS: Elements and edges must never overlap
- Follow ALL 7 Edge Routing Rules for arrow positioning
- Use waypoints to route edges AROUND obstacles
- Use different exitY/entryY values for multiple edges between same nodes
`
// Extended additions (~2600 tokens) - appended for models with 4000 token cache minimum
// Total EXTENDED_SYSTEM_PROMPT = ~4400 tokens
const EXTENDED_ADDITIONS = `
## Extended Tool Reference
### display_diagram Details
**VALIDATION RULES** (XML will be rejected if violated):
1. Generate ONLY mxCell elements - wrapper tags and root cells are added automatically
2. All mxCell elements must be siblings - never nested inside other mxCell elements
3. Every mxCell needs a unique id attribute (start from "2")
4. Every mxCell needs a valid parent attribute (use "1" for top-level, or container-id for grouped)
5. Edge source/target attributes must reference existing cell IDs
6. Escape special characters in values: &lt; for <, &gt; for >, &amp; for &, &quot; for "
**Example with swimlanes and edges** (generate ONLY this - no wrapper tags):
\`\`\`xml
<mxCell id="lane1" value="Frontend" style="swimlane;" vertex="1" parent="1">
<mxGeometry x="40" y="40" width="200" height="200" as="geometry"/>
</mxCell>
<mxCell id="step1" value="Step 1" style="rounded=1;" vertex="1" parent="lane1">
<mxGeometry x="20" y="60" width="160" height="40" as="geometry"/>
</mxCell>
<mxCell id="lane2" value="Backend" style="swimlane;" vertex="1" parent="1">
<mxGeometry x="280" y="40" width="200" height="200" as="geometry"/>
</mxCell>
<mxCell id="step2" value="Step 2" style="rounded=1;" vertex="1" parent="lane2">
<mxGeometry x="20" y="60" width="160" height="40" as="geometry"/>
</mxCell>
<mxCell id="edge1" style="edgeStyle=orthogonalEdgeStyle;endArrow=classic;" edge="1" parent="1" source="step1" target="step2">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
\`\`\`
### append_diagram Details
**WHEN TO USE:** Only call this tool when display_diagram output was truncated (you'll see an error message about truncation).
**CRITICAL RULES:**
1. Do NOT include any wrapper tags - just continue the mxCell elements
2. Continue from EXACTLY where your previous output stopped
3. Complete the remaining mxCell elements
4. If still truncated, call append_diagram again with the next fragment
**Example:** If previous output ended with \`<mxCell id="x" style="rounded=1\`, continue with \`;" vertex="1">...\` and complete the remaining elements.
### edit_diagram Details
**CRITICAL RULES:**
- Copy-paste the EXACT search pattern from the "Current diagram XML" in system context
- Do NOT reorder attributes or reformat - the attribute order in draw.io XML varies and you MUST match it exactly
- Only include the lines that are changing, plus 1-2 surrounding lines for context if needed
- Break large changes into multiple smaller edits
- Each search must contain complete lines (never truncate mid-line)
- First match only - be specific enough to target the right element
**Input Format:**
\`\`\`json
{
"edits": [
{
"search": "EXACT lines copied from current XML (preserve attribute order!)",
"replace": "Replacement lines"
}
]
}
\`\`\`
## edit_diagram Best Practices
### Core Principle: Unique & Precise Patterns
Your search pattern MUST uniquely identify exactly ONE location in the XML. Before writing a search pattern:
1. Review the "Current diagram XML" in the system context
2. Identify the exact element(s) to modify by their unique id attribute
3. Include enough context to ensure uniqueness
### Pattern Construction Rules
**Rule 1: Always include the element's id attribute**
\`\`\`json
{"search": "<mxCell id=\\"node5\\"", "replace": "<mxCell id=\\"node5\\" value=\\"New Label\\""}
\`\`\`
**Rule 2: Include complete XML elements when possible**
\`\`\`json
{
"search": "<mxCell id=\\"3\\" value=\\"Old\\" style=\\"rounded=1;\\" vertex=\\"1\\" parent=\\"1\\">\\n <mxGeometry x=\\"100\\" y=\\"100\\" width=\\"120\\" height=\\"60\\" as=\\"geometry\\"/>\\n</mxCell>",
"replace": "<mxCell id=\\"3\\" value=\\"New\\" style=\\"rounded=1;\\" vertex=\\"1\\" parent=\\"1\\">\\n <mxGeometry x=\\"100\\" y=\\"100\\" width=\\"120\\" height=\\"60\\" as=\\"geometry\\"/>\\n</mxCell>"
}
\`\`\`
**Rule 3: Preserve exact whitespace and formatting**
Copy the search pattern EXACTLY from the current XML, including leading spaces, line breaks (\\n), and attribute order.
### Good vs Bad Patterns
**BAD:** \`{"search": "value=\\"Label\\""}\` - Too vague, matches multiple elements
**BAD:** \`{"search": "<mxCell value=\\"X\\" id=\\"5\\""}\` - Reordered attributes won't match
**GOOD:** \`{"search": "<mxCell id=\\"5\\" parent=\\"1\\" style=\\"...\\" value=\\"Old\\" vertex=\\"1\\">"}\` - Uses unique id with full context
### ⚠️ JSON Escaping (CRITICAL)
Every double quote inside JSON string values MUST be escaped with backslash:
- **CORRECT:** \`"x=\\"100\\" y=\\"200\\""\` - both quotes escaped
- **WRONG:** \`"x=\\"100\\" y="200\\""\` - missing backslash causes JSON parse error!
### Error Recovery
If edit_diagram fails with "pattern not found":
1. **First retry**: Check attribute order - copy EXACTLY from current XML
2. **Second retry**: Expand context - include more surrounding lines
3. **Third retry**: Try matching on just \`<mxCell id="X"\` prefix + full replacement
4. **After 3 failures**: Fall back to display_diagram to regenerate entire diagram
## Edge Examples
### Two edges between same nodes (CORRECT - no overlap):
\`\`\`xml
<mxCell id="e1" value="A to B" style="edgeStyle=orthogonalEdgeStyle;exitX=1;exitY=0.3;entryX=0;entryY=0.3;endArrow=classic;" edge="1" parent="1" source="a" target="b">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
<mxCell id="e2" value="B to A" style="edgeStyle=orthogonalEdgeStyle;exitX=0;exitY=0.7;entryX=1;entryY=0.7;endArrow=classic;" edge="1" parent="1" source="b" target="a">
<mxGeometry relative="1" as="geometry"/>
</mxCell>
\`\`\`
### Edge with single waypoint (simple detour):
\`\`\`xml
<mxCell id="edge1" style="edgeStyle=orthogonalEdgeStyle;exitX=0.5;exitY=1;entryX=0.5;entryY=0;endArrow=classic;" edge="1" parent="1" source="a" target="b">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="300" y="150"/>
</Array>
</mxGeometry>
</mxCell>
\`\`\`
### Edge with waypoints (routing AROUND obstacles) - CRITICAL PATTERN:
**Scenario:** Hotfix(right,bottom) → Main(center,top), but Develop(center,middle) is in between.
**WRONG:** Direct diagonal line crosses over Develop
**CORRECT:** Route around the OUTSIDE (go right first, then up)
\`\`\`xml
<mxCell id="hotfix_to_main" style="edgeStyle=orthogonalEdgeStyle;exitX=0.5;exitY=0;entryX=1;entryY=0.5;endArrow=classic;" edge="1" parent="1" source="hotfix" target="main">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="750" y="80"/>
<mxPoint x="750" y="150"/>
</Array>
</mxGeometry>
</mxCell>
\`\`\`
This routes the edge to the RIGHT of all shapes (x=750), then enters Main from the right side.
**Key principle:** When connecting distant nodes diagonally, route along the PERIMETER of the diagram, not through the middle where other shapes exist.`
// Extended system prompt = DEFAULT + EXTENDED_ADDITIONS
export const EXTENDED_SYSTEM_PROMPT = DEFAULT_SYSTEM_PROMPT + EXTENDED_ADDITIONS
// Model patterns that require extended prompt (4000 token cache minimum)
// These patterns match Opus 4.5 and Haiku 4.5 model IDs
const EXTENDED_PROMPT_MODEL_PATTERNS = [
"claude-opus-4-5", // Matches any Opus 4.5 variant
"claude-haiku-4-5", // Matches any Haiku 4.5 variant
]
/**
* Get the appropriate system prompt based on the model ID and style preference
* Uses extended prompt for Opus 4.5 and Haiku 4.5 which have 4000 token cache minimum
* @param modelId - The AI model ID from environment
* @param minimalStyle - If true, removes style instructions to save tokens
* @returns The system prompt string
*/
export function getSystemPrompt(
modelId?: string,
minimalStyle?: boolean,
): string {
const modelName = modelId || "AI"
let prompt: string
if (
modelId &&
EXTENDED_PROMPT_MODEL_PATTERNS.some((pattern) =>
modelId.includes(pattern),
)
) {
console.log(
`[System Prompt] Using EXTENDED prompt for model: ${modelId}`,
)
prompt = EXTENDED_SYSTEM_PROMPT
} else {
console.log(
`[System Prompt] Using DEFAULT prompt for model: ${modelId || "unknown"}`,
)
prompt = DEFAULT_SYSTEM_PROMPT
}
// Add style instructions based on preference
// Minimal style: prepend instruction at START (more prominent)
// Normal style: append at end
if (minimalStyle) {
console.log(`[System Prompt] Minimal style mode ENABLED`)
prompt = MINIMAL_STYLE_INSTRUCTION + prompt
} else {
prompt += STYLE_INSTRUCTIONS
}
return prompt.replace("{{MODEL_NAME}}", modelName)
}

39
lib/token-counter.ts Normal file
View File

@@ -0,0 +1,39 @@
/**
* Token counting utilities using js-tiktoken
*
* Uses cl100k_base encoding (GPT-4) which is close to Claude's tokenization.
* This is a pure JavaScript implementation, no WASM required.
*/
import { encodingForModel } from "js-tiktoken"
import { DEFAULT_SYSTEM_PROMPT, EXTENDED_SYSTEM_PROMPT } from "./system-prompts"
const encoder = encodingForModel("gpt-4o")
/**
* Count the number of tokens in a text string
* @param text - The text to count tokens for
* @returns The number of tokens
*/
export function countTextTokens(text: string): number {
return encoder.encode(text).length
}
/**
* Get token counts for the system prompts
* Useful for debugging and optimizing prompt sizes
* @returns Object with token counts for default and extended prompts
*/
export function getSystemPromptTokenCounts(): {
default: number
extended: number
additions: number
} {
const defaultTokens = countTextTokens(DEFAULT_SYSTEM_PROMPT)
const extendedTokens = countTextTokens(EXTENDED_SYSTEM_PROMPT)
return {
default: defaultTokens,
extended: extendedTokens,
additions: extendedTokens - defaultTokens,
}
}

110
lib/use-file-processor.tsx Normal file
View File

@@ -0,0 +1,110 @@
"use client"
import { useState } from "react"
import { toast } from "sonner"
import {
extractPdfText,
extractTextFileContent,
isPdfFile,
isTextFile,
MAX_EXTRACTED_CHARS,
} from "@/lib/pdf-utils"
export interface FileData {
text: string
charCount: number
isExtracting: boolean
}
/**
* Hook for processing file uploads, especially PDFs and text files.
* Handles text extraction, character limit validation, and cleanup.
*/
export function useFileProcessor() {
const [files, setFiles] = useState<File[]>([])
const [pdfData, setPdfData] = useState<Map<File, FileData>>(new Map())
const handleFileChange = async (newFiles: File[]) => {
setFiles(newFiles)
// Extract text immediately for new PDF/text files
for (const file of newFiles) {
const needsExtraction =
(isPdfFile(file) || isTextFile(file)) && !pdfData.has(file)
if (needsExtraction) {
// Mark as extracting
setPdfData((prev) => {
const next = new Map(prev)
next.set(file, {
text: "",
charCount: 0,
isExtracting: true,
})
return next
})
// Extract text asynchronously
try {
let text: string
if (isPdfFile(file)) {
text = await extractPdfText(file)
} else {
text = await extractTextFileContent(file)
}
// Check character limit
if (text.length > MAX_EXTRACTED_CHARS) {
const limitK = MAX_EXTRACTED_CHARS / 1000
toast.error(
`${file.name}: Content exceeds ${limitK}k character limit (${(text.length / 1000).toFixed(1)}k chars)`,
)
setPdfData((prev) => {
const next = new Map(prev)
next.delete(file)
return next
})
// Remove the file from the list
setFiles((prev) => prev.filter((f) => f !== file))
continue
}
setPdfData((prev) => {
const next = new Map(prev)
next.set(file, {
text,
charCount: text.length,
isExtracting: false,
})
return next
})
} catch (error) {
console.error("Failed to extract text:", error)
toast.error(`Failed to read file: ${file.name}`)
setPdfData((prev) => {
const next = new Map(prev)
next.delete(file)
return next
})
}
}
}
// Clean up pdfData for removed files
setPdfData((prev) => {
const next = new Map(prev)
for (const key of prev.keys()) {
if (!newFiles.includes(key)) {
next.delete(key)
}
}
return next
})
}
return {
files,
pdfData,
handleFileChange,
setFiles, // Export for external control (e.g., clearing files)
}
}

247
lib/use-quota-manager.tsx Normal file
View File

@@ -0,0 +1,247 @@
"use client"
import { useCallback, useMemo } from "react"
import { toast } from "sonner"
import { QuotaLimitToast } from "@/components/quota-limit-toast"
import { STORAGE_KEYS } from "@/lib/storage"
export interface QuotaConfig {
dailyRequestLimit: number
dailyTokenLimit: number
tpmLimit: number
}
export interface QuotaCheckResult {
allowed: boolean
remaining: number
used: number
}
/**
* Hook for managing request/token quotas and rate limiting.
* Handles three types of limits:
* - Daily request limit
* - Daily token limit
* - Tokens per minute (TPM) rate limit
*
* Users with their own API key bypass all limits.
*/
export function useQuotaManager(config: QuotaConfig): {
hasOwnApiKey: () => boolean
checkDailyLimit: () => QuotaCheckResult
checkTokenLimit: () => QuotaCheckResult
checkTPMLimit: () => QuotaCheckResult
incrementRequestCount: () => void
incrementTokenCount: (tokens: number) => void
incrementTPMCount: (tokens: number) => void
showQuotaLimitToast: () => void
showTokenLimitToast: (used: number) => void
showTPMLimitToast: () => void
} {
const { dailyRequestLimit, dailyTokenLimit, tpmLimit } = config
// Check if user has their own API key configured (bypass limits)
const hasOwnApiKey = useCallback((): boolean => {
const provider = localStorage.getItem(STORAGE_KEYS.aiProvider)
const apiKey = localStorage.getItem(STORAGE_KEYS.aiApiKey)
return !!(provider && apiKey)
}, [])
// Generic helper: Parse count from localStorage with NaN guard
const parseStorageCount = (key: string): number => {
const count = parseInt(localStorage.getItem(key) || "0", 10)
return Number.isNaN(count) ? 0 : count
}
// Generic helper: Create quota checker factory
const createQuotaChecker = useCallback(
(
getTimeKey: () => string,
timeStorageKey: string,
countStorageKey: string,
limit: number,
) => {
return (): QuotaCheckResult => {
if (hasOwnApiKey())
return { allowed: true, remaining: -1, used: 0 }
if (limit <= 0) return { allowed: true, remaining: -1, used: 0 }
const currentTime = getTimeKey()
const storedTime = localStorage.getItem(timeStorageKey)
let count = parseStorageCount(countStorageKey)
if (storedTime !== currentTime) {
count = 0
localStorage.setItem(timeStorageKey, currentTime)
localStorage.setItem(countStorageKey, "0")
}
return {
allowed: count < limit,
remaining: limit - count,
used: count,
}
}
},
[hasOwnApiKey],
)
// Generic helper: Create quota incrementer factory
const createQuotaIncrementer = useCallback(
(
getTimeKey: () => string,
timeStorageKey: string,
countStorageKey: string,
validateInput: boolean = false,
) => {
return (tokens: number = 1): void => {
if (validateInput && (!Number.isFinite(tokens) || tokens <= 0))
return
const currentTime = getTimeKey()
const storedTime = localStorage.getItem(timeStorageKey)
let count = parseStorageCount(countStorageKey)
if (storedTime !== currentTime) {
count = 0
localStorage.setItem(timeStorageKey, currentTime)
}
localStorage.setItem(countStorageKey, String(count + tokens))
}
},
[],
)
// Check daily request limit
const checkDailyLimit = useMemo(
() =>
createQuotaChecker(
() => new Date().toDateString(),
STORAGE_KEYS.requestDate,
STORAGE_KEYS.requestCount,
dailyRequestLimit,
),
[createQuotaChecker, dailyRequestLimit],
)
// Increment request count
const incrementRequestCount = useMemo(
() =>
createQuotaIncrementer(
() => new Date().toDateString(),
STORAGE_KEYS.requestDate,
STORAGE_KEYS.requestCount,
false,
),
[createQuotaIncrementer],
)
// Show quota limit toast (request-based)
const showQuotaLimitToast = useCallback(() => {
toast.custom(
(t) => (
<QuotaLimitToast
used={dailyRequestLimit}
limit={dailyRequestLimit}
onDismiss={() => toast.dismiss(t)}
/>
),
{ duration: 15000 },
)
}, [dailyRequestLimit])
// Check daily token limit
const checkTokenLimit = useMemo(
() =>
createQuotaChecker(
() => new Date().toDateString(),
STORAGE_KEYS.tokenDate,
STORAGE_KEYS.tokenCount,
dailyTokenLimit,
),
[createQuotaChecker, dailyTokenLimit],
)
// Increment token count
const incrementTokenCount = useMemo(
() =>
createQuotaIncrementer(
() => new Date().toDateString(),
STORAGE_KEYS.tokenDate,
STORAGE_KEYS.tokenCount,
true, // Validate input tokens
),
[createQuotaIncrementer],
)
// Show token limit toast
const showTokenLimitToast = useCallback(
(used: number) => {
toast.custom(
(t) => (
<QuotaLimitToast
type="token"
used={used}
limit={dailyTokenLimit}
onDismiss={() => toast.dismiss(t)}
/>
),
{ duration: 15000 },
)
},
[dailyTokenLimit],
)
// Check TPM (tokens per minute) limit
const checkTPMLimit = useMemo(
() =>
createQuotaChecker(
() => Math.floor(Date.now() / 60000).toString(),
STORAGE_KEYS.tpmMinute,
STORAGE_KEYS.tpmCount,
tpmLimit,
),
[createQuotaChecker, tpmLimit],
)
// Increment TPM count
const incrementTPMCount = useMemo(
() =>
createQuotaIncrementer(
() => Math.floor(Date.now() / 60000).toString(),
STORAGE_KEYS.tpmMinute,
STORAGE_KEYS.tpmCount,
true, // Validate input tokens
),
[createQuotaIncrementer],
)
// Show TPM limit toast
const showTPMLimitToast = useCallback(() => {
const limitDisplay =
tpmLimit >= 1000 ? `${tpmLimit / 1000}k` : String(tpmLimit)
toast.error(
`Rate limit reached (${limitDisplay} tokens/min). Please wait 60 seconds before sending another request.`,
{ duration: 8000 },
)
}, [tpmLimit])
return {
// Check functions
hasOwnApiKey,
checkDailyLimit,
checkTokenLimit,
checkTPMLimit,
// Increment functions
incrementRequestCount,
incrementTokenCount,
incrementTPMCount,
// Toast functions
showQuotaLimitToast,
showTokenLimitToast,
showTPMLimitToast,
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,7 +1,8 @@
import type { NextConfig } from "next";
import type { NextConfig } from "next"
const nextConfig: NextConfig = {
/* config options here */
};
/* config options here */
output: "standalone",
}
export default nextConfig;
export default nextConfig

4603
package-lock.json generated

File diff suppressed because it is too large Load Diff

View File

@@ -1,53 +1,88 @@
{
"name": "next-ai-draw-io",
"version": "0.2.0",
"version": "0.4.1",
"license": "Apache-2.0",
"private": true,
"scripts": {
"dev": "next dev --turbopack --port 6002",
"build": "next build",
"start": "next start --port 6001",
"lint": "next lint"
"lint": "biome lint .",
"format": "biome check --write .",
"check": "biome ci",
"prepare": "husky"
},
"dependencies": {
"@ai-sdk/amazon-bedrock": "^3.0.52",
"@ai-sdk/amazon-bedrock": "^3.0.70",
"@ai-sdk/anthropic": "^2.0.44",
"@ai-sdk/azure": "^2.0.69",
"@ai-sdk/deepseek": "^1.0.30",
"@ai-sdk/google": "^2.0.0",
"@ai-sdk/openai": "^2.0.19",
"@ai-sdk/react": "^2.0.22",
"@ai-sdk/react": "^2.0.107",
"@aws-sdk/credential-providers": "^3.943.0",
"@langfuse/client": "^4.4.9",
"@langfuse/otel": "^4.4.4",
"@langfuse/tracing": "^4.4.9",
"@next/third-parties": "^16.0.6",
"@openrouter/ai-sdk-provider": "^1.2.3",
"@opentelemetry/exporter-trace-otlp-http": "^0.208.0",
"@opentelemetry/sdk-trace-node": "^2.2.0",
"@radix-ui/react-collapsible": "^1.1.12",
"@radix-ui/react-dialog": "^1.1.6",
"@radix-ui/react-label": "^2.1.8",
"@radix-ui/react-scroll-area": "^1.2.3",
"@radix-ui/react-select": "^2.2.6",
"@radix-ui/react-slot": "^1.1.2",
"@radix-ui/react-switch": "^1.2.6",
"@radix-ui/react-tooltip": "^1.1.8",
"@vercel/analytics": "^1.5.0",
"@radix-ui/react-use-controllable-state": "^1.2.2",
"@xmldom/xmldom": "^0.9.8",
"ai": "^5.0.89",
"base-64": "^1.0.0",
"class-variance-authority": "^0.7.1",
"clsx": "^2.1.1",
"js-tiktoken": "^1.0.21",
"jsdom": "^26.0.0",
"jsonrepair": "^3.13.1",
"lucide-react": "^0.483.0",
"next": "15.2.3",
"motion": "^12.23.25",
"next": "^16.0.7",
"ollama-ai-provider-v2": "^1.5.4",
"pako": "^2.1.0",
"react": "^19.0.0",
"react-dom": "^19.0.0",
"prism-react-renderer": "^2.4.1",
"react": "^19.1.2",
"react-dom": "^19.1.2",
"react-drawio": "^1.0.3",
"react-icons": "^5.5.0",
"react-markdown": "^10.1.0",
"react-resizable-panels": "^3.0.6",
"remark-gfm": "^4.0.1",
"sonner": "^2.0.7",
"tailwind-merge": "^3.0.2",
"tailwindcss-animate": "^1.0.7",
"unpdf": "^1.4.0",
"zod": "^4.1.12"
},
"lint-staged": {
"*.{js,ts,jsx,tsx,json,css}": [
"biome check --write --no-errors-on-unmatched",
"biome check --no-errors-on-unmatched"
]
},
"devDependencies": {
"@anthropic-ai/tokenizer": "^0.0.4",
"@biomejs/biome": "2.3.8",
"@tailwindcss/postcss": "^4",
"@tailwindcss/typography": "^0.5.19",
"@types/node": "^20",
"@types/pako": "^2.0.3",
"@types/react": "^19",
"@types/react-dom": "^19",
"eslint": "9.39.1",
"eslint-config-next": "16.0.5",
"husky": "^9.1.7",
"lint-staged": "^16.2.7",
"tailwindcss": "^4",
"typescript": "^5"
}

View File

@@ -1,5 +1,5 @@
const config = {
plugins: ["@tailwindcss/postcss"],
};
plugins: ["@tailwindcss/postcss"],
}
export default config;
export default config

View File

@@ -0,0 +1,65 @@
Here is an extended summary of the paper **"Chain-of-Thought Prompting Elicits Reasoning in Large Language Models"** by Jason Wei, et al. This detailed overview covers the background, methodology, extensive experimental results, emergent properties, and qualitative analysis found in the study.
### **1. Introduction and Motivation**
The paper addresses a significant limitation in Large Language Models (LLMs): while scaling up model size (increasing parameters) has revolutionized performance on standard NLP tasks, it has not proven sufficient for challenging logical tasks such as arithmetic, commonsense, and symbolic reasoning.
Traditional techniques to solve these problems fell into two camps:
1. **Finetuning:** Training models manually with large datasets of explanations (expensive and task-specific).
2. **Standard Few-Shot Prompting:** Providing input-output pairs (e.g., Question $\rightarrow$ Answer) without explaining *how* the answer was derived. This often fails on multi-step problems.
The authors introduce **Chain-of-Thought (CoT) Prompting**, a simple method that combines the strengths of both approaches. It leverages the model's existing capabilities to generate natural language rationales without requiring any model parameter updates (finetuning).
### **2. Methodology: What is Chain-of-Thought?**
The core innovation is changing the structure of the "exemplars" (the few-shot examples included in the prompt).
* **Standard Prompting:** The model is shown a question and an immediate answer.
* *Q: Roger has 5 balls. He buys 2 cans of 3 balls. How many now?*
* *A: 11.*
* **Chain-of-Thought Prompting:** The model is shown a question, followed by a series of intermediate natural language reasoning steps that lead to the answer.
* *A: Roger started with 5 balls. 2 cans of 3 tennis balls each is 6 tennis balls. 5 + 6 = 11. The answer is 11.*
By interacting with the model using this format, the LLM learns to generate its own "thought process" for new, unseen questions. This allows the model to decompose complex problems into manageable intermediate steps.
### **3. Experimental Setup**
The researchers evaluated CoT prompting on several large language models, including **GPT-3 (175B)**, **LaMDA (137B)**, **PaLM (540B)**, **UL2 (20B)**, and **Codex**. They tested across three distinct domains of reasoning:
* **Arithmetic Reasoning:** Using benchmarks like **GSM8K** (math word problems), **SVAMP**, **ASDiv**, **AQuA**, and **MAWPS**.
* **Commonsense Reasoning:** Using datasets like **CSQA**, **StrategyQA**, **Date Understanding**, and **Sports Understanding**.
* **Symbolic Reasoning:** Using tasks like **Last Letter Concatenation** and **Coin Flip** tracking (determining if a coin is heads or tails after a sequence of flips).
### **4. Key Findings and Results**
#### **Arithmetic Reasoning**
The results on math word problems were striking. Standard prompting struggled significantly, often exhibiting a flat scaling curve (performance didn't improve much even as models got bigger).
* **Performance Jump:** On the difficult **GSM8K** benchmark, **PaLM 540B** with CoT prompting achieved **56.9%** accuracy, compared to just 17.9% with standard prompting.
* **Surpassing State-of-the-Art:** PaLM 540B with CoT outperformed a previously finetuned GPT-3 model (55%), establishing a new state-of-the-art without needing a training set.
* **Calculator Integration:** The authors noted that some errors were simple calculation mistakes in otherwise correct logic. By hooking the CoT output into an external Python calculator, accuracy on GSM8K rose further to **58.6%**.
#### **Commonsense Reasoning**
CoT prompting improved performance on tasks requiring background knowledge and physical intuition.
* **StrategyQA:** PaLM 540B achieved **75.6%** accuracy via CoT, beating the prior state-of-the-art (69.4%).
* **Sports Understanding:** The model achieved **95.4%** accuracy, surpassing the performance of an unaided sports enthusiast (84%).
* The gains were minimal on CSQA, likely because many questions in that dataset did not require multi-step logic.
#### **Symbolic Reasoning and Generalization**
A unique strength of CoT was enabling **Out-of-Domain (OOD) Generalization**.
* In the **Coin Flip** task, the models were given examples with only 2 flips. However, using CoT, the models could successfully track coins flipped 3 or 4 times.
* Standard prompting failed completely on these longer sequences, while CoT allowed the model to repeat the logical steps as many times as necessary to reach the solution.
### **5. Emergent Ability of Scale**
One of the paper's most critical insights is that CoT reasoning is an **emergent ability** that depends on model size.
* **Small Models (<10B parameters):** CoT prompting provided **no benefit** and often hurt performance. Small models produced fluent but illogical chains of thought (hallucinations) or suffered from repetition.
* **Large Models (~100B+ parameters):** The ability to reason sequentially emerges at this scale. The performance gains from CoT are negligible for small models but increase dramatically for models like GPT-3 (175B) and PaLM (540B).
### **6. Why Does It Work? (Ablation Studies)**
To ensure the improvement was due to the reasoning steps and not other factors, the authors conducted three specific ablations:
1. **Equation Only:** They prompted the model to output just the math equation without words. This performed worse than CoT, suggesting that natural language helps the model "understand" the question semantics.
2. **Variable Compute:** They prompted the model to output dots (...) to consume compute time before answering. This yielded no improvement, proving that the *content* of the reasoning steps matters, not just the extra tokens.
3. **Reasoning After Answer:** They asked the model to give the answer first, then the explanation. This performed about the same as the baseline, proving that the chain of thought must come *before* the answer to guide the model's inference process.
### **7. Error Analysis and Robustness**
The authors manually analyzed errors made by the models.
* **Error Types:** In math problems, errors were categorized as **Semantic Understanding** (misunderstanding the question), **One-Step Missing** (skipping a logical step), or **Calculation Errors**.
* **Impact of Scale:** Scaling from PaLM 62B to PaLM 540B significantly reduced semantic and missing-step errors, confirming that larger models are better at logic, not just memorization.
* **Robustness:** The method proved robust to different annotators (different people writing the prompts) and different specific examples, though, like all prompting, different prompt styles did result in some variance.
### **Conclusion**
The paper establishes Chain-of-Thought prompting as a powerful paradigm for unlocking the reasoning potential of Large Language Models. By simply asking the model to "show its work," researchers can elicit complex logical behaviors that were previously thought to require specialized architectures or extensive finetuning. The work highlights that reasoning is an emergent capability of sufficiently large language models.

BIN
public/favicon.ico Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 15 KiB

View File

@@ -0,0 +1,4 @@
<svg xmlns="http://www.w3.org/2000/svg" width="140" height="36" viewBox="0 0 140 36">
<rect width="140" height="36" rx="8" fill="#6366f1"/>
<text x="70" y="24" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif" font-size="15" font-weight="600" fill="white" text-anchor="middle">🚀 Live Demo</text>
</svg>

After

Width:  |  Height:  |  Size: 340 B

View File

@@ -1,27 +1,33 @@
{
"compilerOptions": {
"target": "ES2017",
"lib": ["dom", "dom.iterable", "esnext"],
"allowJs": true,
"skipLibCheck": true,
"strict": true,
"noEmit": true,
"esModuleInterop": true,
"module": "esnext",
"moduleResolution": "bundler",
"resolveJsonModule": true,
"isolatedModules": true,
"jsx": "preserve",
"incremental": true,
"plugins": [
{
"name": "next"
}
"compilerOptions": {
"target": "ES2017",
"lib": ["dom", "dom.iterable", "esnext"],
"allowJs": true,
"skipLibCheck": true,
"strict": true,
"noEmit": true,
"esModuleInterop": true,
"module": "esnext",
"moduleResolution": "bundler",
"resolveJsonModule": true,
"isolatedModules": true,
"jsx": "react-jsx",
"incremental": true,
"plugins": [
{
"name": "next"
}
],
"paths": {
"@/*": ["./*"]
}
},
"include": [
"next-env.d.ts",
"**/*.ts",
"**/*.tsx",
".next/types/**/*.ts",
".next/dev/types/**/*.ts"
],
"paths": {
"@/*": ["./*"]
}
},
"include": ["next-env.d.ts", "**/*.ts", "**/*.tsx", ".next/types/**/*.ts"],
"exclude": ["node_modules"]
"exclude": ["node_modules"]
}

12
vercel.json Normal file
View File

@@ -0,0 +1,12 @@
{
"functions": {
"app/api/chat/route.ts": {
"memory": 512,
"maxDuration": 120
},
"app/api/**/route.ts": {
"memory": 256,
"maxDuration": 10
}
}
}