* refactor(canvas): decouple the canvas from the chat engine
The canvas components will be reused by the MCP server's browser shell,
which has no chat and no Next.js. They now read everything they need
from small contexts and the canvas store instead of the chat engine:
- components/canvas/versions-context.tsx: VersionsProvider and
useVersionsContext give the version cards, the strip and the compare
dialog a VersionsSource (versions, onCanvasId, undoneId, isBusy,
canUndo, canRedo, restore, undo, redo). The chat fills it from the
versions store and the engine in components/chat/chat-versions.tsx.
- stores/canvas-store.ts: isBusy and busyReason; the chat engine sets
them while a turn runs, SelectionAsk and the version UI read them.
- components/canvas/version-card.tsx: the visual version card and the
thumbnail, out of tool-activity.tsx; the chat's card composes it and
adds its "Show XML" link and code panel.
- compare-dialog.tsx and version-strip.tsx move to components/canvas;
the strip takes the minimum number of versions to show as a prop, the
chat panel computes it from the cards it has.
- components/canvas/locale-context.tsx: LocaleProvider and useLocale,
fed by the [lang] layout; CanvasStage no longer uses next/navigation.
The overlay (SelectionAsk) is a slot and the wait for the saved
language is a prop, so the shell can leave both out.
- lib/version-text.ts: describeChanges is now describeChangeSummary, so
it can be imported next to the MCP core's describeChanges.
* test(unit): canvas import boundary and the shared version card
The boundary test scans the canvas modules' imports and bundles the canvas components with esbuild; both fail on anything from components/chat, the tool handlers, next/navigation, next/font, next/script, next/headers or a server-only module. The card test renders VersionCard and VersionStrip on a fake VersionsSource.
* refactor(scripts): pin the draw.io release in one json file and share the zip reader
* feat(mcp-server): bundle a trimmed draw.io into dist/drawio at build time
scripts/fetch-drawio.mjs downloads the pinned draw.war into a cache
(DRAWIO_WAR_CACHE, or DRAWIO_WAR for a local file), checks its SHA-256
and extracts the files named in drawio-files.txt plus a LICENSE with the
Apache-2.0 text. scripts/check-drawio-files.mjs drives the embedded
editor in headless Chromium from a full copy, records every requested
file and checks or rewrites (--update) the list.
* feat(mcp-server): serve the bundled draw.io same-origin behind an api token
GET /drawio/<path> serves dist/drawio with a MIME table, a day of
caching and nosniff; paths are normalized and never reach WEB-INF or
META-INF. The preview embeds /drawio/index.html when the copy exists and
DRAWIO_BASE_URL is unset, else the external draw.io as before (and
start_session says so). Every /api request must carry the per-process
X-Drawio-Token the page gets in its HTML; pages send
frame-ancestors 'self' and nosniff.
* build(mcp-server): require the bundled draw.io in the package and cap the tarball at 40 MB
* fix(mcp-server): review fixes for the preview token and the draw.io static files
- The preview page fetches a fresh copy of itself and retries once when an
API request is refused with 403: another MCP process, with its own token,
now answers on this port, and the recovery logic (recoverState) needs its
polls to go through. The page is sent with Cache-Control: no-store.
- draw.io files are served with an ETag and Cache-Control: no-cache instead
of a 24 hour max-age: their names do not change between versions, so a
package upgrade must reach the browser on the next preview. HEAD and
If-None-Match (304) are answered.
- The file read stream goes through stream.pipeline, so a read error no
longer ends the MCP process and a client that leaves mid-download no
longer leaks the file handle.
* fix(mcp-server): review fixes for the draw.io file list and its guard
- The list now ships the templates of Insert > Template, the PlantUML
parser of Insert > Advanced and the template dialog's icon: the menu
items were shown but failed with 404s. 15.1 MB packed; the tarball cap
goes from 40 MB to 20 MB, where it still catches a list that grew by a
whole js/ directory.
- The guard drives both features, fails when the copy it runs against
lacks files the editor asked for or when an export does not answer,
checks that the copy is the pinned draw.io version, and runs in CI
(npm run check-drawio, one E2E shard) so a list regression cannot reach
a release.
* docs: review fixes for the draw.io version pointer and the preview's embedding
- The offline deployment guides point at packages/mcp-server/src/
drawio-version.json for the draw.war version; scripts/fetch-drawio.mjs
no longer names it.
- The MCP README says what the bundled draw.io copy includes, and that the
preview page cannot be shown inside an editor's built-in browser since it
is served with frame-ancestors 'self'.
* refactor(canvas): review fixes for the busy flag, the comparison note and the bundle size check
- busyReason had no reader: the chat engine sets isBusy through the store's
generic set, like every other flag.
- The isSameDocument comment says the two "same document" rules disagree
in both directions, so neither is a subset of the other.
- The import boundary test bundles minified with one pako (the MCP core
resolves its own copy) and caps the canvas core at 160 KB (134 KB now).
* fix(mcp-server): name exports at random and stamp the draw.io version into ETags
- Export requests carried a per-process counter. The preview page retries a
result refused with 403 against the process that took over the port, so a
late result of the old process's export could be taken for the new
process's export with the same number. The id is a random UUID now.
- The ETag of a bundled draw.io file now starts with the version the fetch
script stamps into dist/drawio/.version. An install that keeps the
archive's dates (npm does not) would otherwise answer 304 for a changed
file of the same size after an upgrade.
* docs(mcp-server): mermaid is in the bundled draw.io, which has no image proxy
- Mermaid's converter is in js/extensions.min.js, which the editor loads at
startup, so the README no longer lists it as left out; the org chart
layout (js/orgchart.min.js) is the example of a feature that is.
- Note that the bundled copy, like the web app's, has no /drawio/proxy:
images from other websites, including those in some templates, are left
out of exports and thumbnails. The guard's comment names this as a known
limitation it does not cover.
* refactor(canvas): getDrawioSrc takes the editor's source, and CanvasStage can be given one
* feat(mcp-server): canvas shell bundled from the web app's canvas with esbuild
The shell (packages/mcp-server/shell) renders DrawioFrame inside the
shared providers without the chat: its config comes from
window.__MCP_CONFIG__, the four dictionaries ship in the bundle, the
theme is kept under an mcp: localStorage key, and system fonts stand in
for the web fonts. scripts/build-shell.mjs bundles it into dist/shell
with the root's esbuild and Tailwind; check-package caps shell.js at
1.5 MB and shell.css at 300 KB.
* feat(mcp-server): serve the canvas shell at /shell/ behind DRAWIO_PREVIEW_UI
GET /shell/ fills the page template's {{CONFIG_JSON}} with the session,
the API token, where draw.io comes from and the host's editor settings,
with the classic page's security headers; /shell/<file> serves the built
files like the draw.io copy. start_session opens the shell when
DRAWIO_PREVIEW_UI=shell; the classic page stays the default.
* test(mcp-server): open the built shell with playwright
npm run test:e2e starts dist/index.js over the MCP SDK with
DRAWIO_PREVIEW_UI=shell, opens the URL start_session returns in headless
Chromium, and checks that the bundled draw.io loads without console
errors and that the theme toggle switches draw.io in place.
* ci: install the root dependencies before building the mcp package
The shell build needs the root's esbuild, Tailwind and React.
* feat(canvas): DiagramProvider exports with options and shows a transient document
requestExport(request, timeoutMs) runs one draw.io export with the
request's own parameters (a PNG's pageId and width) and resolves with its
answer (data, or xml for format "xml"); null when draw.io does not answer.
showTransient(xml) puts a document on the canvas for an export only: it is
not recorded, its autosaves are ignored, and the next loadDiagram brings
the user back to the page they were on. The MCP's canvas shell answers the
server's export requests with both.
* feat(mcp-server): the shell syncs with the server like the classic page
shell/mcp-sync-core.ts ports preview.js's protocol without the page or
React: polling GET /api/state every 2 s with the state id and version,
pushes of the user's edits with their base version and state, 409 and 413
handling, recovery of a recreated session (the tab's copy goes to History),
the token refresh after a 403, sync and export requests (PNG by page id,
SVG of another page through a transient one-page projection with autosave
ignored, the 600 ms wait for icons), thumbnails for History, and History
reads and restores that name the state the list belongs to.
A server write goes on the canvas as one undo step (loadDiagram commit,
with the change marked) when decideLoad finds it changes only the page on
screen, keeps the file variables and needs no load-only setting; the other
pages are compared with the MCP core's contentFingerprint. draw.io's own
re-serialized copy of such a write is recognized with isSameDocument and
not pushed back as an edit.
shell/use-mcp-sync.ts mounts the sync once inside DiagramProvider, reads
draw.io's autosave and load messages, and reports the connection state,
which the shell's status bar shows; notices come as toasts from the
dictionaries.
* test(mcp-server): port the preview protocol tests to the sync core
The recovery, thumbnail, size limit and History cases of
tests/unit/mcp-preview-recovery.test.ts, driven against mcp-sync-core.ts
with a stubbed server and canvas, plus the stale 409 recovery, an export
request answered once, a projection that ignores edits and restores the
document, the thumbnail of a write loaded before draw.io was ready, writes
taken in place as commits (draw.io's own copy is not pushed), and the
commit-or-load decision table of decideLoad.
* test(mcp-server): e2e of the shell's sync with the server
The shell connects, shows what create_new_diagram draws, marks an
edit_diagram change and takes it back with one Ctrl+Z (which get_diagram
then reflects), pushes a shape inserted in the editor, serves
screenshot_diagram, and shows another page only for its SVG export before
the user's page comes back.
* fix(mcp-server): review fixes for the shell's sync core
A recovery loads the server's diagram in full and waits for draw.io's
load report, as the classic page does: whether the canvas takes a write
in place is decided inside DiagramProvider, and an autosave of the
canvas being replaced went to the server as an edit when the decision
was wrong. A paper size change alone is pushed again: autosaves are
compared as documents only for draw.io's own copy of a committed write.
A projection stays on screen until draw.io reports the restore load
(5 s at most), the whole export ends in 15 s even when its result POST
hangs, and an export in flight is dropped when draw.io starts over. A
refused poll shows the tab offline. With several pages and the page on
screen unknown, a write loads in full. hasLoadOnlySettings and the
highlight of an AI change are shared with the editor bridge.
* fix(mcp-server): review fixes for the preview URL and the theme menu
The stalled-tab note names the page start_session opened (the shell
with DRAWIO_PREVIEW_UI=shell) instead of the classic page. With the
shell not built, start_session opens the classic page and the server
says so at startup, as it does for a missing dist/drawio. BROWSER=none
skips the system browser, as Vite and CRA dev servers do: the e2e tests
set it. Without a fixed DRAWIO_UI, the shell's draw.io URL carries
themes=1 so Extras > Theme is offered, as on the classic page.
* fix(mcp-server): review fixes for the shell's build
tsc -p shell checked nothing: the inherited exclude dropped every shell
file; the shell's tsconfig now excludes node_modules only. Tailwind scans
app-toaster.tsx too, so the sync notices look like the web app's toasts.
The bundle keeps its dependencies' license comments (esbuild appends them
at the end). postcss, which build-shell.mjs loads from the root, is a
declared devDependency.
* ci: run the shell's browser tests on the packed tarball, publish on shared code changes
The e2e job's first shard builds the MCP package, packs it, installs the
tarball in an empty directory and runs packages/mcp-server/tests/e2e
against the installed dist/index.js (MCP_SERVER_ENTRY), so a shell that
builds but fails in the browser, or a tarball missing a runtime
dependency, fails the PR. Its traces go with the Playwright report. The
publish workflow also triggers on the web app code the shell is bundled
from (components, contexts, hooks, lib, stores, globals.css) and the
root lockfile.
* fix(canvas): ignore autosaves of the document a pending full load replaces
An autosave that arrives after loadDiagram sent a full load, and before draw.io reports it, is of the canvas being replaced (draw.io reports each load, in order). Until now it overwrote chartXMLRef: in the MCP shell, a late autosave of a one-page projection replaced the cached document, and a remount of an external draw.io then put the projection on the canvas as the whole document.
* fix(mcp-server): review fixes for the shell's commits and export timeout
A write taken in place that leaves the canvas as it is gets no autosave from draw.io, so the marker for its copy stayed armed and ate the user's next paper-size edit: arm it only when the write changed the document. The 15 s export timeout now retires the export's sequence number, so a result POST that answers late no longer restores the document a second time.
* ci: create the tarball directory before npm pack
npm pack does not create its --pack-destination; on a fresh runner the step failed with ENOENT before the packed shell tests ran.
* feat(mcp-server): history entries name their maker, and /api/history the entry on the canvas
Each History entry records who made it (a user edit, a recovering tab's
copy, a restored copy; none for the AI's writes), when, and its page
count. GET /api/history now also returns each entry's XML, the first entry
with the same content (a restore adds a copy of an older one), the state
the ids belong to, and the entry whose content the canvas shows, by the
rule restore_version uses. The shell's version cards read these; the
classic page keeps using index, id and svg.
* feat(mcp-server): version cards in the shell from the server's History
The shell's VersionsSource (shell/node-versions-source.ts) reads GET
/api/history through the sync and shows one version per distinct content
(a restore's copy folds into the entry it copies; the blank page is none),
numbered as they first appeared, with what changed since the one before.
The entry the server says is on the canvas marks the version; undo and
redo of the newest version restore the one before it, or it again, through
POST /api/restore, so hand edits stay as versions of their own. The cards
sit in a panel beside the canvas (toggle in the header); the strip and
Compare come from the shared canvas components. The sync tells listeners
once per server version, and when History changed without one, so the
list is read again only then.
* test(mcp-server): e2e of the shell's version cards
In a session of its own: three writes give three cards; undoing the newest
restores the one before on the server too, redo brings it back, a hand edit
turns the undo into a restore, and restoring the first version keeps that
edit as a version of its own, which restore_version also finds.
* feat(mcp-server): get_selection reads the cells the user selected in the shell
The server asks the preview tab for the selection the way it asks for an export (a random request id in GET /api/state, the answer in a POST with that id, 10 s to answer). The shell answers through the editor bridge with each cell's id, label, an edge's ends, a shape's geometry and the page on screen; without a same-origin editor it says so, and the tool names the external draw.io. The classic page cannot answer, so the tool says that at once. screenshot_diagram's description now says a page selector renders that page without changing the page on screen (PNG exports by pageId already did).
* test(mcp-server): selection requests, get_selection texts, and page screenshots that leave the view alone
Unit tests for the request plumbing (random id, one answer, timeout), the tool's texts (cells, nothing selected, external draw.io) and the shell's side (one answer per request, after draw.io is up, never during a projection); the wiring test starts a session with BROWSER=none to see the no-tab message. Shell e2e: cells selected in the editor reach get_selection; a PNG of another page differs from the page on screen, which stays.
* feat(mcp-server): the canvas shell is the default preview, the classic page behind DRAWIO_PREVIEW_UI=classic
* docs: the MCP preview is the web app's canvas; get_selection, DRAWIO_BASE_URL and the classic page
* chore(mcp-server): version 0.4.0; check-package requires the draw.io version stamp
* ci: an mcp-shell job builds the package, runs its tests, the draw.io file guard and the shell e2e
* fix(mcp-server): review fixes for the version cards and History
The shell's version cards (shell/node-versions-source.ts):
- a version's change and undo target are the state it replaced, the
History entry right before its first copy, not the card before it: after
a restore those differ, and undo went to the wrong version (and not
where restore_version steps_back=1 goes)
- a card restores the newest copy of its content, as restore_version
does, so page settings the user changed (a "user" copy) are kept
- a blank page after a drawing is a clear of the canvas, a version of its
own; only the blank page before any drawing is hidden
- numbers and changes are keyed by content, not by the first copy's id,
so a version keeps them when its first copy drops out of the server's
20-entry buffer; the caches start over for another server state (the
process restarted: entry ids name other content)
The server's History (src/history.ts):
- firstCopyIds compares each entry with the first of every group only: a
bare model matches any page name, so "same content" is not transitive,
and a card could show one document and restore another
- the time and pages fields had no reader; pages parsed every XML once
more on every write
Reading History (src/http-server.ts, shell/mcp-sync-core.ts):
- GET /api/state and a push's answer carry a History key (entry count,
newest id, the entry on the canvas); the shell reads History again only
when it changes, so a hand edit no longer downloads every entry's XML and
thumbnail
- a failed History read is told again at the next poll
- a History list from a state the poll has not seen yet is dropped
* fix(mcp-server): review fixes for get_selection
Server (src/http-server.ts, src/index.ts, src/selection.ts,
src/new-diagram.ts):
- overlapping get_selection calls take turns (readSelection, one slot per
session as the export slot) instead of replacing each other's request,
which left one of them with a false "tab not in front" timeout
- an answer is taken only while a request is pending (both ids undefined
compared equal)
- the tool sees the page start_session actually opens: with dist/shell
missing the classic page is in use, which never answers, so the tool
says so instead of timing out
- the result lists at most 100 cells and counts the rest: a whole large
diagram selected would fill the model's context
- every <diagram> the model sends without an id gets one, so the page id
the shell reports exists in the server's document
Shell (shell/mcp-sync-core.ts, shell/use-mcp-sync.ts,
contexts/diagram-context.tsx, lib/drawio/editor-bridge.ts):
- an answer whose POST failed is sent again at the next poll
- no answer while a full load has yet to reach the editor: it still shows
the previous document, whose cells and pages the answer would name
- a hidden tab (the same session open twice) answers a poll later, so the
tab in front answers first; alone, it still answers within the timeout
- a cell's container is reported by the model's isLayer, not by comparing
with the default parent, which is the group the user entered
* fix(mcp-server): review fixes for the shell page
- a download button in the header opens the web app's export dialog
(.drawio, .png, .svg, .drawio.svg), which the classic page had and the
shell lacked when it became the default
- the shell asks draw.io for the custom library menu (libraries=1), as
the classic page did; the web app keeps libraries=0
- the newest card no longer shows "Rendering preview" for good: the sync
takes the thumbnail of a diagram the server recovered from its file
(saved without pictures) while the canvas kept it, and of a write whose
picture was skipped because an edit came first, once the canvas shows
the write again
- e2e: the get_selection test covers a shape in a container the user
entered; a download test saves a .drawio file
* ci: the version bump reminder watches every path publish-mcp.yml does, and compares the version field
The reminder step left out the root lockfile and scripts/, which change
what is bundled into the shell and trigger the publish workflow; and it
took any change of packages/mcp-server/package.json (a dependency bump)
for a version bump. It now compares the version field between the two
commits.
* fix(mcp-server): get_selection lists every selected id, with detail for the first 100
* fix(mcp-server): a hidden tab's delayed selection answer checks the canvas again first
* fix(mcp-server): the version cards keep only the current History in memory, picture the newest copy, and keep a clear the buffer scrolled to
* fix(mcp-server): save who made each History entry with it, so the cards look the same after a restart
* fix(mcp-server): the download waits for a page export's projection to end, and the dialog's styles are bundled
* fix(mcp-server): bundle open.html and js/open.js, the picker of Open Library from > Browser
* fix(mcp-server): the shell shows the versions once, in the side panel
* test(mcp-server): get_selection wiring test expects the classic page's answer when the shell is not built
16 KiB
Next AI Draw.io MCP Server
MCP (Model Context Protocol) server that enables AI agents like Claude Desktop and Cursor to generate and edit draw.io diagrams with real-time browser preview.
Self-contained: the embedded HTTP server and the draw.io editor are in the package, so after install the preview works offline.
Quick Start
{
"mcpServers": {
"drawio": {
"command": "npx",
"args": ["@next-ai-drawio/mcp-server@latest"]
}
}
}
Installation
Claude Desktop
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"drawio": {
"command": "npx",
"args": ["@next-ai-drawio/mcp-server@latest"]
}
}
}
VS Code
Add to your VS Code settings (.vscode/mcp.json in workspace or user settings):
{
"mcpServers": {
"drawio": {
"command": "npx",
"args": ["@next-ai-drawio/mcp-server@latest"]
}
}
}
Cursor
Add to Cursor MCP config (~/.cursor/mcp.json):
{
"mcpServers": {
"drawio": {
"command": "npx",
"args": ["@next-ai-drawio/mcp-server@latest"]
}
}
}
Cline (VS Code Extension)
- Click the MCP Servers icon in Cline's top menu bar
- Select the Configure tab
- Click Configure MCP Servers to edit
cline_mcp_settings.json - Add the drawio server:
{
"mcpServers": {
"drawio": {
"command": "npx",
"args": ["@next-ai-drawio/mcp-server@latest"]
}
}
}
Claude Code CLI
claude mcp add drawio -- npx @next-ai-drawio/mcp-server@latest
Other MCP Clients
Use the standard MCP configuration with:
- Command:
npx - Args:
["@next-ai-drawio/mcp-server@latest"]
Usage
-
Restart your MCP client after updating config
-
Ask the AI to create a diagram:
"Create a flowchart showing user authentication with login, MFA, and session management"
"Read docs/architecture.md and draw the system as a diagram"
-
The diagram appears in your browser in real-time!
Features
- Real-time Preview: Diagrams appear and update in your browser as the AI creates them. The preview is the web app's canvas: a change the AI makes to the page on screen is outlined, and one Ctrl+Z takes it back
- Selection: Select shapes in the preview and say "move these" or "connect this box to that one"; the AI reads what you selected with
get_selection - Drawing Rules: The AI gets the same layout, edge and style rules as the web app, plus the shape library docs (AWS, Azure, GCP, Kubernetes, Cisco and more), so it uses real icon names instead of guessing
- Self-check: The AI can take a screenshot of the rendered diagram (
screenshot: trueoncreate_new_diagramoredit_diagram, orscreenshot_diagram) and fix overlapping shapes or edges that cross shapes - Draw from Your Files: ask the AI to draw from a document, image or web page; it reads the source with the host's own tools and draws. Existing .drawio files open with load_diagram
- Edit Support: Modify existing diagrams with natural language instructions. If any change in an edit fails, nothing is written and the AI gets the reason and the current page XML
- Your Edits Are Kept: Changes you make in the browser are read before the AI edits again. If the AI overwrites a change you were still making, your version is saved in History
- Version History: The last 20 versions appear as cards next to the canvas, with a thumbnail and what changed. Restore one, undo and redo the latest, or compare a version with the canvas; or ask the AI to undo (
restore_version) - Export: Save as
.drawio,.png,.svg, or.drawio.svg(an SVG with the diagram embedded, which draw.io andload_diagramcan open again) throughexport_diagram, or from the download button in the preview's header - Multi-page: List, add, rename, and delete pages, and edit any page
- Auto-save: Each session's diagram is saved to
~/.next-ai-drawio/<session-id>.drawio, and its last 20 versions to<session-id>.history.json, so the diagram, History and undo survive a restart of the MCP client - Custom Instructions: keep your own drawing rules in
~/.next-ai-drawio/instructions.md(for example "Always draw in minimal style"); they are appended to the drawing guide on every call - Themes and Dark Mode: Pick a draw.io theme under Extras > Theme; the button at the top right switches the page and the editor between light and dark (the system setting until you choose).
DRAWIO_UI,DRAWIO_DARKandDRAWIO_LANGfix them - Self-contained: The draw.io editor ships inside the package and is served from the preview's own origin, so after install the preview works offline.
DRAWIO_BASE_URLpoints it at another draw.io instead
Available Tools
| Tool | Description |
|---|---|
start_session |
Opens browser with real-time diagram preview; the result includes the drawing rules. Pass session_id to continue a saved diagram |
list_saved_diagrams |
List the auto-saved diagrams of earlier sessions, newest first, with their pages |
get_drawing_guide |
Return the drawing rules again, for example after a long conversation was compacted, or, with topic, a short XML reference for tables, layers or groups |
get_shape_library |
Return the shapes and icon styles of a library such as aws4, azure2, or kubernetes |
create_new_diagram |
Create a new diagram from XML; a plain list of mxCell elements is enough |
load_diagram |
Load a .drawio or .drawio.svg file into the session, from a path on disk or from its xml content (handles compressed files) |
edit_diagram |
Edit diagram by ID-based operations (update/add/delete cells); all or nothing |
get_diagram |
Get the current diagram XML, including your edits in the browser |
get_selection |
Return the shapes and edges you selected in the preview (ids, labels, positions), so the AI can act on "these" |
screenshot_diagram |
Return a PNG of a page so the AI can check the rendered diagram |
export_diagram |
Save diagram to a .drawio, .png, .svg, or .drawio.svg file |
list_pages |
List every page (tab) with id, name, index, and cell count |
add_page |
Append a new page without touching existing ones |
rename_page |
Rename a page |
delete_page |
Delete a page (refuses to delete the last one) |
restore_version |
Undo: put an earlier version from History back on the canvas; the current one is kept, so a redo is possible |
Continue a Diagram Later
After every change, the diagram is saved as a normal .drawio file in ~/.next-ai-drawio/, named after the session id (for example mcp-mgd0a1b2-x7k2p1.drawio). start_session tells the AI both the id and the file path.
To continue a diagram in a later conversation (for example after claude --resume, or in a new chat), the AI calls start_session with session_id set to that id: the same preview URL opens, the saved diagram is shown, and auto-save keeps writing to the same file. If the id is no longer in the conversation, list_saved_diagrams returns every saved diagram, newest first, with the names and cell counts of its pages, so you can ask for "the architecture diagram from yesterday" and let the AI pick it. load_diagram with the file path still works too, and you can open the file in draw.io yourself.
History comes back with the diagram: the last 20 versions are saved next to it in <session-id>.history.json (without thumbnails). When the AI continues the session with start_session and its session_id, or a preview tab is still open after a restart, the version cards show them again and restore_version can undo to them.
The newest 50 files are kept, each with its History file. Set DRAWIO_DATA_DIR to use another folder, or to off to turn auto-save off.
Custom Instructions
To give the AI your own drawing rules, write them in ~/.next-ai-drawio/instructions.md as Markdown, for example "Always draw in minimal style" or "Label every edge". The file is appended to the drawing guide under a ## Custom Instructions heading each time the guide is returned (start_session, get_drawing_guide and the diagram-workflow prompt), so edits apply without restarting your MCP client. Only the first 5000 characters are used.
DRAWIO_DATA_DIR changes the folder the file is read from; with DRAWIO_DATA_DIR=off the default folder is still used, since the file is only read. The text reaches the model as it is, so keep it to drawing rules you trust.
How It Works
┌─────────────────┐ stdio ┌─────────────────┐
│ Claude Desktop │ <───────────> │ MCP Server │
│ (AI Agent) │ │ (this package) │
└─────────────────┘ └────────┬────────┘
│
┌────────▼────────┐
│ Embedded HTTP │
│ Server (:6002) │
└────────┬────────┘
│
┌────────▼────────┐
│ User's Browser │
│ (draw.io embed) │
└─────────────────┘
- MCP Server receives tool calls from Claude via stdio
- Embedded HTTP Server serves the preview page and the draw.io editor, and handles state
- Browser shows real-time diagram updates via polling
The server listens on 127.0.0.1 only. Every /api request must carry a token that the server generates at start and writes into the preview page, so another website open in the same browser cannot read or change the diagram; the preview page also refuses to be embedded in other pages. Opening the preview URL in another browser on the same machine works, since the page comes with the token.
Configuration
| Variable | Default | Description |
|---|---|---|
PORT |
6002 |
Port for the embedded HTTP server |
DRAWIO_BASE_URL |
unset (the bundled draw.io) | An external draw.io for the preview, such as a self-hosted instance (see below). The page cannot reach into an editor from another origin, so with it the AI's changes are not outlined, Ctrl+Z does not take them back and get_selection cannot read the selection. The version cards and everything else keep working. |
DRAWIO_DATA_DIR |
~/.next-ai-drawio |
Folder for the auto-saved .drawio files and your instructions.md. Set to off to turn auto-save off (instructions.md is then read from the default folder). |
DRAWIO_LANG |
unset | Language of the draw.io editor and of the preview's own texts. Unset, the preview's texts follow the browser language, and draw.io chooses: English on the bundled copy until the user picks one under Extras > Language, the browser language on embed.diagrams.net. A code such as en, zh, zh-tw, ja or de fixes both and hides that submenu. |
DRAWIO_UI |
unset | draw.io theme. Unset, the user picks one under Extras > Theme and draw.io remembers it. kennedy, atlas, dark, min, sketch or simple fixes the theme and hides that menu. |
DRAWIO_DARK |
auto |
Dark mode of the preview and the editor until the user switches it with the button at the top right (the browser remembers that choice): auto follows the system, 1 starts dark, 0 starts light. |
DRAWIO_PREVIEW_UI |
shell |
The preview page start_session opens. classic opens the previous preview page (see Classic preview page). |
DRAWIO_AUTO_SCREENSHOT |
unset | Set to true to attach a screenshot to every create_new_diagram and edit_diagram result, so the AI checks each drawing. Costs 2 to 10 s per call; the preview tab must be open and in front. A call can still pass screenshot: false. |
BROWSER |
unset | Set to none and start_session does not open the preview in the system browser (the convention of Vite and Create React App dev servers); the result still names the URL. |
DEBUG |
unset | Set to true to log debug messages to stderr. |
Private Deployment (Self-hosted draw.io)
For security-sensitive environments that require private deployment of draw.io:
{
"mcpServers": {
"drawio": {
"command": "npx",
"args": ["@next-ai-drawio/mcp-server@latest"],
"env": {
"DRAWIO_BASE_URL": "https://drawio.your-company.com"
}
}
}
}
You can deploy your own draw.io instance using the official Docker image:
docker run -d -p 8080:8080 jgraph/drawio
Then set DRAWIO_BASE_URL=http://localhost:8080 (or your server's URL). The preview page loads nothing else from the internet, so with a local draw.io it works offline. One exception: shapes from the Material Design library show icons from fonts.gstatic.com.
With an external draw.io the preview cannot reach into the editor (browsers keep pages and frames from other origins apart), so the AI's changes are not outlined on the canvas, Ctrl+Z does not take them back (use the version cards or restore_version), and get_selection cannot read what you selected. start_session tells the AI about this mode.
Without DRAWIO_BASE_URL the preview uses the trimmed draw.io copy bundled with this package. It has the editor, the shape libraries, the templates, PlantUML and Mermaid; features that load more code when used (the org chart layout, for example) are not included. Set DRAWIO_BASE_URL to a full draw.io if you need them.
Like the web app's bundled copy, it has no image proxy (/drawio/proxy): images from other websites, including the avatars in the Org Chart and Mind Map templates and the Arista icons in the network templates, show on the canvas only when online and are left out of PNG and SVG exports and version thumbnails. Insert such images from a file instead, or set DRAWIO_BASE_URL to a full draw.io (the jgraph/drawio Docker image includes the proxy). See Offline deployment.
Classic preview page
DRAWIO_PREVIEW_UI=classic opens the preview page of the previous releases: draw.io with a History dialog and a Download button, without the outline of the AI's changes, Ctrl+Z for them, the version cards, Compare and get_selection. It stays for one more release and is then removed.
Troubleshooting
Port already in use
If port 6002 is in use, the server will automatically try the next available port (up to 6020).
Or set a custom port:
{
"mcpServers": {
"drawio": {
"command": "npx",
"args": ["@next-ai-drawio/mcp-server@latest"],
"env": { "PORT": "6003" }
}
}
}
"No active session"
Call start_session first to open the browser window.
Browser not updating
Check that the browser URL has the ?mcp= query parameter. The MCP session ID connects the browser to the server.
Screenshot or PNG/SVG export times out
PNG and SVG files are rendered by draw.io in the preview tab. Browsers slow down tabs that stay in the background, so the tab may not answer in time. Bring the preview tab to the front and try again.
Preview is blank inside VS Code Simple Browser
The preview page refuses to be embedded in other pages (it sends Content-Security-Policy: frame-ancestors 'self'), so in-editor browsers such as VS Code Simple Browser, Cursor's built-in browser or a port forward's "Preview in Editor" show a blank page. Open the Browser URL from the start_session result in a regular browser (Chrome, Edge, Firefox, Safari) instead.
License
Apache-2.0