wren MCP server
ai.wrenapp/wrenwrenapp.aistreamable-httpv1.0.0Scanned
Ask your AI to write in your Wren notebook: notes, graphs, flowcharts and formulas as real ink.
13 tools: 4 can delete or cancel things, 2 do something that can't be undone, 1 change things and 6 only read.
- Reads private data · 5
- Reads text others write · 7
- Can send data out · 0
- 10 findings
Tools listed without sign-in. Remote posture
With Anlyon
4 tools here can delete or cancel things and 2 do something that can't be undone. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the wren MCP server can do
The wren MCP server is published by wrenapp.ai, a domain the MCP Registry verified. It has 13 tools. It can delete or cancel things (4 tools: edit_page, update_note, write_notebook_page and plan_notebook), do things that cannot be undone, such as sending, paying or publishing (create_handwritten_note and add_page) and create or change data (1 tool: connect_account). 6 tools only read.
5 tools read private data (render_note_preview, list_notes, get_note and 2 more); 7 tools bring in text other people can write (create_handwritten_note, render_note_preview, list_recent_notebooks and 4 more).
The fixed checks matched 10 findings (credentials-mention and annotation-conflict). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 14, 2026, version 1.0.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
13 tools
Deletes or cancels 4Claude Code: deny
edit_pageWrite handwriting onto an EXISTING page that Wren wrote.Reads text others writeNo findings
Why it is in this group
- The name "edit_page" says "edit": a word for a change.
- The input addresses people (input.graphSpec.shade.to): what it sends cannot be taken back.
- The server marks it destructive (destructiveHint: true).
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Write handwriting onto an EXISTING page that Wren wrote. mode 'append' (default) keeps the page's ink and adds the new ink below it; mode 'replace' erases the page's ink and draws the new ink. Only pages get_note reports as editable (Wren-written and not changed on the iPad since) can be edited — the user's own handwriting is never modified; use add_page for those notes. Pick the page by pageNumber (1-based) or pageId; omitted = last page. If an append has no room on the page, Wren puts it on a new page at the end of the note and says so.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: true
- idempotentHint: false
plan_notebookPlan a 1-30 page handwritten Wren notebook and write its cover and contents (page 1; a plan over 8 pages gets a cover page and contents pages).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it destructive (destructiveHint: true).
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Plan a 1-30 page handwritten Wren notebook and write its cover and contents (page 1; a plan over 8 pages gets a cover page and contents pages). Pass noteId to extend a plan and update the contents. Then write each page via write_notebook_page with target.noteId and planSlot, one call per slot. Same topic as an earlier note → pass its noteId, not a new note. Optional: part (topic group, e.g. "Lecture 3") per page; coverFigure = one small sketch of the topic.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: true
- idempotentHint: false
update_noteChange a note's title and/or subject.1 finding
Why it is in this group
- The name "update_note" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Findings
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Description, as the server wrote it Change a note's title and/or subject. Does not touch its pages.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
- idempotentHint: true
write_notebook_pageWrite a handwritten study page in the user's Wren notebook (same topic -> add/replace via target.noteId, not a new note).Reads text others writeNo findings
Why it is in this group
- The name "write_notebook_page" says "write": a word for a change.
- The server marks it destructive (destructiveHint: true).
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Write a handwritten study page in the user's Wren notebook (same topic -> add/replace via target.noteId, not a new note). Write a FULL page (not for verbatim): 110-140 words in 6-8 blocks (a table or figure = ~40 words), a heading per part, 3-6 key terms marked (underline or highlight, never a whole callout; a mark copies exact words of its own block), every number+unit and formula marked as math, one accent, at most 1 short marginNote, and the takeaway in a callout (key_answer, key_idea) or summary. Pair two <=12-word callouts with layoutHint:"side_by_side". Editable ink in Wren's handwriting font, not the user's own handwriting. Recipes: worked_example = problem callout (box) + 3-4 steps (1-2 sentences + equation each) + Check heading+paragraph + common_mistake & key_answer callouts. concept_summary = definition callout + boxed equation or figure (table/flowchart, required) + 2 headings with 2-sentence paragraphs + key_idea callout. comparison_sheet = 1-sentence intro + 5-6 row table or bar_chart + heading+paragraph on why it matters + common_mistake & key_idea callouts (side_by_side). process_walkthrough = overview + 4-6 steps, each with label (stage name) + 1-2 sentences, a flowchart of the stages (cycle: last edge back to the first) + key_idea or summary. revision_sheet = 2-3 headings with concise paragraphs/equations + common_mistake & remember callouts (side_by_side) + 4-item checklist. verbatim = text the user gave you to copy: their words unchanged and once, in paragraph/heading blocks, nothing added.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: true
- idempotentHint: false
Can't be undone 2Claude Code: ask
add_pageAdd a NEW handwritten page at the end of an existing note (same drawing inputs as create_handwritten_note).Reads text others writeNo findings
Why it is in this group
- The name "add_page" says "add": a word for a change.
- The input addresses people (input.graphSpec.shade.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Add a NEW handwritten page at the end of an existing note (same drawing inputs as create_handwritten_note). Returns which note and page the ink went to and the note's outline; use it whenever the user stays on the note's topic.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: true
- idempotentHint: false
create_handwritten_noteCreate a NEW note in the user's Wren notebook with a handwritten page (pressure-faithful PencilKit strokes) drawn from a `figure{kind, content}`.Reads text others writeNo findings
Why it is in this group
- The name "create_handwritten_note" says "create": a word for a change.
- The input addresses people (input.graphSpec.shade.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Create a NEW note in the user's Wren notebook with a handwritten page (pressure-faithful PencilKit strokes) drawn from a `figure{kind, content}`. The result has a link (noteUrl) that opens the note on any device — always give it to the user — and the note also syncs to the Wren iPad app. To add to a note that already exists use add_page (new page) or edit_page (same page) with its noteId — find it with list_notes or list_recent_notebooks. Same topic as an earlier note → add a page to it, not a new note. A figure is required: graphs (graph_2d, content.prompt = the equation), handwritten text (annotation, content.text = the exact words), formulas, tables, bar charts, timelines, flowcharts, mind maps, concept maps and more — call get_drawing_guide for every kind. Write in the user's language: Wren handwrites Hindi, Arabic, Bengali, Tamil, Telugu, Hebrew, Russian and most other scripts (not yet Chinese / Japanese / Korean) — never transliterate. A concept map reads best with at most 10 concepts and 11 links per page; a bigger one is split over pages automatically (links between pages are drawn on both with a page reference).
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: true
- idempotentHint: false
Changes things 1Claude Code: ask
connect_accountLink the user's Wren account so authored notes land in their notebook.3 findings
Why it is in this group
- The description's first verb is "link": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Info
The description mentions "credential".
What this check looks forMatched text
description: "...notebook. Returns the linked uid when a credential is present; otherwise explains that acc..."
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
- Warn
The server marks "connect_account" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "link": a word for a change.
Description, as the server wrote it Link the user's Wren account so authored notes land in their notebook. Returns the linked uid when a credential is present; otherwise explains that account linking must be completed first.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
Only reads 6Show 6 read-only tools
get_drawing_guideReturns what Wren can draw (every kind), the exact arguments for each with a working example, and the rules for writing into notes.Reads private data1 finding
Why it is in this group
- The name "get_drawing_guide" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "return" and "note": it reads notes.
Findings
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Description, as the server wrote it Returns what Wren can draw (every kind), the exact arguments for each with a working example, and the rules for writing into notes. Call it before your first drawing in a conversation, or whenever unsure which kind or fields to use. Read-only; needs no account.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
get_noteShow one note's title, subject, page size and its pages (number, id, ink stroke count, whether Wren can add ink to it).Reads private data1 finding
Why it is in this group
- The name "get_note" says "get": a word for reading.
What it can reach
- private-dataThe name "get_note" says "get" and "note": it reads notes.
- private-dataThe description's first sentence says "show" and "note": it reads notes.
Findings
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Description, as the server wrote it Show one note's title, subject, page size and its pages (number, id, ink stroke count, whether Wren can add ink to it). Use before edit_page to pick the page.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
get_profileReturn the stable Wren account id (Firebase uid) and basic profile of the currently linked account.1 finding
Why it is in this group
- The name "get_profile" says "get": a word for reading.
Findings
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Description, as the server wrote it Return the stable Wren account id (Firebase uid) and basic profile of the currently linked account. Call this after connect_account to confirm linkage.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
list_notesList the user's Wren notes, most recently edited first, with id, title, subject and page count.Reads private data1 finding
Why it is in this group
- The name "list_notes" says "list": a word for reading.
What it can reach
- private-dataThe name "list_notes" says "list" and "note": it reads notes.
- private-dataThe description's first sentence says "list" and "note": it reads notes.
Findings
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Description, as the server wrote it List the user's Wren notes, most recently edited first, with id, title, subject and page count. Use it to find the noteId before add_page, edit_page, get_note or update_note.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
list_recent_notebooksThe user's latest notes, newest first: id, title, subject, pages, last edit, the outline of pages Wren wrote, and the note this chat has been writing to.Reads private dataReads text others write1 finding
Why it is in this group
- The name "list_recent_notebooks" says "list": a word for reading.
What it can reach
- private-dataThe description's first sentence says "note": it reads notes.
- untrusted-inputThe description's first sentence says "chat": it reads messages, which other people can write.
Findings
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Description, as the server wrote it The user's latest notes, newest first: id, title, subject, pages, last edit, the outline of pages Wren wrote, and the note this chat has been writing to. Use it when the user returns to a topic, then add pages to that noteId.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
render_note_previewRender a pressure-faithful SVG preview of a handwritten note's strokes (the same VectorStroke array create_handwritten_note produces), framed at the page size, with an Open in Wren…Reads private dataReads text others write1 finding
Why it is in this group
- The name "render_note_preview" says "preview": a word for reading.
What it can reach
- private-dataThe name "render_note_preview" says "preview" and "note": it reads notes.
- private-dataThe description's first sentence says "note": it reads notes.
- untrusted-inputThe description's first sentence says "url": it reads web pages, which anyone can write.
Findings
- Info
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Description, as the server wrote it Render a pressure-faithful SVG preview of a handwritten note's strokes (the same VectorStroke array create_handwritten_note produces), framed at the page size, with an Open in Wren button (pass the note link as openInNoteNestUrl). Returns an inline widget.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged outbound.
- Reads private data5
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write7
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 21, 2026, 19:20 UTC, unchanged since Nov 14, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 4
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://mcp.wrenapp.ai/.well-known/oauth-protected-resource/mcp
- Authorization servers
- https://mcp.wrenapp.ai
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- yes
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 4 tools that change things (3 destructive, 1 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://mcp.wrenapp.ai/mcp: tools/list returned 8 tools to an anonymous client, including write_notebook_page (destructive), plan_notebook (destructive), draw_figure (destructive), connect_account (write)
Findings
Matches of fixed checks, with the text they matched.
- Warnconnect_account
The server marks "connect_account" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "link": a word for a change.
- Infoget_profile
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
- Infoconnect_account
The description mentions "credential".
What this check looks forMatched text
description: "...notebook. Returns the linked uid when a credential is present; otherwise explains that acc..."
- Infoconnect_account
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
- Infolist_notes
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
Show 4 more findingsShow fewer
- Infoget_note
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
- Infoupdate_note
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
The input schema at properties.context.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.context.description: "...es, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http wren https://mcp.wrenapp.ai/mcpclaude mcp add --transport http wren https://mcp.wrenapp.ai/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__wren__get_profile",
"mcp__wren__render_note_preview",
"mcp__wren__list_notes",
"mcp__wren__get_note",
"mcp__wren__list_recent_notebooks",
"mcp__wren__get_drawing_guide"
],
"ask": [
"mcp__wren__connect_account",
"mcp__wren__create_handwritten_note",
"mcp__wren__add_page"
],
"deny": [
"mcp__wren__edit_page",
"mcp__wren__update_note",
"mcp__wren__write_notebook_page",
"mcp__wren__plan_notebook"
]
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as wren. Review it before you use it.
Capability tags behind 10 rules
- mcp__wren__create_handwritten_note untrusted-input
- mcp__wren__render_note_preview private-data, untrusted-input
- mcp__wren__list_notes private-data
- mcp__wren__get_note private-data
- mcp__wren__list_recent_notebooks private-data, untrusted-input
- mcp__wren__add_page untrusted-input
- mcp__wren__edit_page untrusted-input
- mcp__wren__get_drawing_guide private-data
- mcp__wren__write_notebook_page untrusted-input
- mcp__wren__plan_notebook untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/ai.wrenapp/wren/verify?hash=37405722695060a14cd6d217973bc77fc02185aa68ffae0db75b9ef9649ccffa" | jq -e '.changed == false'Pinned hash 374057226950, from Nov 14, 2026, 06:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/ai.wrenapp/wren/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/ai.wrenapp/wren)Source: MCP Registry. Synced Nov 14, 2026, 06:17 UTC.
wren MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the wren MCP server safe?
- No scan can promise that. What its definitions show: 13 tools, 7 of which can change, send or delete something, and 10 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the wren MCP server delete data?
- Yes. 4 tools can delete, cancel or revoke something: edit_page, update_note, write_notebook_page and plan_notebook.
Can the wren MCP server do things that cannot be undone?
- Yes. 2 tools can send, pay, publish or merge: create_handwritten_note and add_page.
What tools does the wren MCP server have?
- 13 tools: 4 destructive, 2 irreversible, 1 write, 6 read. edit_page, update_note, write_notebook_page, plan_notebook, create_handwritten_note, add_page, connect_account, get_profile and 5 more.
Does the wren MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the wren MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the wren MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 14, 2026.