# ai.wrenapp/wren (remote · mcp.wrenapp.ai)

Ask your AI to write in your Wren notebook: notes, graphs, flowcharts and formulas as real ink.

- Trust score: 66/100 (medium)
- Change this week: 0
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `mcp.wrenapp.ai`: 66/100 (this document), [markdown](https://verifymcp.io/servers/ai-wrenapp-wren/mcp.md), [page](https://verifymcp.io/servers/ai-wrenapp-wren/mcp)

## Channel facts

- Endpoint: `https://mcp.wrenapp.ai/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 63/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (write_notebook_page).
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 79/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1482 tokens (~134/item across 11 items; 8 tools + 3 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability check failed: schema churn in the 7 days we've observed: 6 tool removals, 7 breaking changes, 0 auth/transport breaks, 1 additions.
- **Tool Coverage**: 88/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 58% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 8 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 10 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.
  - Supports UI / widget rendering.

## Install

### How do I install the ai.wrenapp/wren MCP server?

ai.wrenapp/wren is a hosted endpoint at https://mcp.wrenapp.ai/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http ai-wrenapp-wren 'https://mcp.wrenapp.ai/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "ai-wrenapp-wren": {
      "url": "https://mcp.wrenapp.ai/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "ai-wrenapp-wren": {
      "type": "http",
      "url": "https://mcp.wrenapp.ai/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.ai-wrenapp-wren]
url = "https://mcp.wrenapp.ai/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-wrenapp-wren": {
      "type": "remote",
      "url": "https://mcp.wrenapp.ai/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-wrenapp-wren --url 'https://mcp.wrenapp.ai/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  ai-wrenapp-wren:
    url: "https://mcp.wrenapp.ai/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "ai-wrenapp-wren": {
      "Transport": "http",
      "Url": "https://mcp.wrenapp.ai/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add ai-wrenapp-wren -t streamable-http -u 'https://mcp.wrenapp.ai/mcp'
```

### Other

```json
{
  "mcpServers": {
    "ai-wrenapp-wren": {
      "type": "http",
      "url": "https://mcp.wrenapp.ai/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-15 (score 66, 0)

- [security regression] Stability: unverified → fail
- [security regression] A breaking change shipped without a version bump: still 0.1.0
- [security regression] Tool “add_page” was removed
- [security regression] Tool “create_handwritten_note” was removed
- [security regression] Tool “edit_page” was removed
- [security regression] Tool “get_drawing_guide” was removed
- [security regression] Tool “render_note_preview” was removed
- [security regression] Tool “update_note” was removed
- [security] The server rewrote its instructions, which are the text every model session reads
- [security] New tool “draw_figure”, which the server declares destructive
- [security] Tool “get_note” rewrote its description, which is the text the model reads
- [security] Tool “get_profile” rewrote its description, which is the text the model reads
- [security] Tool “list_notes” rewrote its description, which is the text the model reads
- [security] Tool “list_recent_notebooks” rewrote its description, which is the text the model reads
- [security] Tool “plan_notebook” rewrote its description, which is the text the model reads
- [security] Tool “write_notebook_page” rewrote its description, which is the text the model reads
- [functional regression] Tool coverage: 83% → 58%
- [functional regression] “connect_account” dropped the required parameter “context”
- [functional regression] “connect_account” dropped the required parameter “llm_model”
- [functional regression] “get_note” dropped the required parameter “context”
- [functional regression] “get_note” dropped the required parameter “llm_model”
- [functional regression] “get_profile” dropped the required parameter “context”
- [functional regression] “get_profile” dropped the required parameter “llm_model”
- [functional regression] “list_notes” dropped the required parameter “context”
- [functional regression] “list_notes” dropped the required parameter “llm_model”
- [functional regression] “list_recent_notebooks” dropped the required parameter “context”
- [functional regression] “list_recent_notebooks” dropped the required parameter “llm_model”
- [functional regression] “plan_notebook” dropped the required parameter “llm_model”
- [functional regression] “write_notebook_page” dropped the required parameter “llm_model”
- [functional improvement] Schema quality: 407 → 134
- [functional] Schema quality: excellent → good
- [cosmetic] “write_notebook_page” added an optional parameter “noteTitle”
- [cosmetic] “plan_notebook” reworded the description of “subject”
- [cosmetic] “write_notebook_page” reworded the description of “accent”
- [cosmetic] “write_notebook_page” reworded the description of “pageKind”
- [cosmetic] “write_notebook_page” reworded the description of “planSlot”
- [cosmetic] “write_notebook_page” reworded the description of “subject”
- [cosmetic] “connect_account” dropped the optional parameter “conversation_id”
- [cosmetic] “get_note” dropped the optional parameter “conversation_id”
- [cosmetic] “get_profile” dropped the optional parameter “conversation_id”
- [cosmetic] “list_notes” dropped the optional parameter “conversation_id”
- [cosmetic] “list_recent_notebooks” dropped the optional parameter “conversation_id”
- [cosmetic] “plan_notebook” dropped the optional parameter “conversation_id”
- [cosmetic] “write_notebook_page” dropped the optional parameter “conversation_id”

### 2026-11-14 (score 66)

First indexed and scored.

## MCP tools (8)

### `write_notebook_page` (~528 tokens)

Write a handwritten Wren notebook page

Writes one handwritten study page into the user's Wren notebook: editable ink, not the user's own handwriting. target.noteId adds it to that note; else a new note. Wren emails the user a link to their first note saved here. A full page (not for verbatim): 110-140 words in 6-8 blocks (a table/figure = ~40 words), a heading per part, 3-6 key terms marked (underline/highlight; a mark's text is exact words of its own block), every number+unit and formula marked as math (its latex keeps every word: 20\ \text{m/s}), one accent, at most 1 short marginNote, the takeaway in a callout (key_answer, key_idea) or summary. layoutHint side_by_side pairs two <=12-word callouts. Unknown fields are refused. Page kinds: 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 table/flowchart + 2 headings with 2-sentence paragraphs + key_idea callout. comparison_sheet = 1-sentence intro + 5-6 row table + 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, optional flowchart showing what the steps don't + key_idea or summary. revision_sheet = 2-3 headings with concise paragraphs/equations + common_mistake & remember callouts (side_by_side) + 4-item checklist. verbatim = the user's own text, unchanged and once, in paragraph/heading blocks, nothing added.

Input parameters:

- `accent` (string): New note only; an existing notebook keeps its accent.
- `blocks` (array, required)
- `context` (object)
- `noteTitle` (string): New title for the note and its cover.
- `pageKind` (string, required): worked_example | concept_summary | comparison_sheet | process_walkthrough | revision_sheet | verbatim: layouts above.
- `planSlot` (integer): Content page k of the plan (cover not a slot); needs target.noteId only.
- `subject` (string): Subject the note is filed under.
- `target` (object)
- `title` (string, required): Page title.

Output parameters:

- `conversationNote` (object)
- `hint` (string): Set when the page came out short: what to add.
- `location` (string)
- `mode` (string)
- `noteId` (string)
- `noteUrl` (string): Private link to the note in the user's Wren account (opens after sign-in).
- `ok` (boolean)
- `outline` (array)
- `pageCount` (number)
- `pageId` (string)
- `pageKind` (string)
- `pageNumber` (number)
- `plan` (object)
- `semanticType` (string)
- `strokeCount` (number)
- `strokeGroupId` (string)
- `title` (string)

### `plan_notebook` (~247 tokens)

Plan a handwritten Wren notebook

Plans a 1-30 page handwritten Wren notebook and writes its cover and contents (page 1; over 8 pages: a cover page and contents pages). The cover is not a slot: slots are the content pages, numbered from 1. Each slot's page is written by write_notebook_page with target.noteId + planSlot, one page per call. With noteId, the plan extends that existing note: new pages become the next slots after the existing ones. Optional per page: part, a topic group such as "Obstetrics" (separate from the page title). Optional coverFigure: one on-topic sketch drawn on the cover. Wren emails the user a link to their first note saved here.

Input parameters:

- `accent` (string)
- `context` (object)
- `coverFigure` (object): e.g. formula {latex}, graph_2d {curves:[{expr}]}, flowchart; drawn in one pen.
- `noteId` (string)
- `pages` (array, required)
- `subject` (string): Subject, printed on the cover and used to file the note.
- `title` (string, required)

Output parameters:

- `conversationNote` (object)
- `location` (string)
- `mode` (string)
- `noteId` (string)
- `noteUrl` (string)
- `ok` (boolean)
- `outline` (array)
- `pageCount` (number)
- `pageId` (string)
- `pageNumber` (number)
- `plan` (object)
- `semanticType` (string)
- `strokeCount` (number)
- `strokeGroupId` (string)
- `title` (string)

### `list_recent_notebooks` (~75 tokens)

Recent Wren notebooks

Returns 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.

Input parameters:

- `cursor` (string): nextCursor from the previous call.
- `limit` (integer): Notes per call (default 5).

Output parameters:

- `conversationNote` (object)
- `nextCursor` (string)
- `notebooks` (array)
- `ok` (boolean)

### `get_note` (~77 tokens)

Get a Wren note

Returns one note's title, subject, page size and its pages (number, id, ink stroke count, whether Wren can add ink to it). Page numbers match target.pageNumber in write_notebook_page and draw_figure.

Input parameters:

- `noteId` (string, required): The note's id, from list_notes or a previous Wren result.

Output parameters:

- `note` (object)
- `ok` (boolean)

### `draw_figure` (~202 tokens)

Draw one figure in Wren

Draws one standalone figure in the user's Wren notebook, large on its own page: a graph, labelled diagram, chart, table, timeline, flowchart, mind map, concept map, number line or shapes. Figure only, no text blocks (write_notebook_page pages can hold figure blocks next to text). No target: a new note. target.noteId: a new page at the end of that note; with target.pageNumber: drawn onto that page if Wren wrote it (mode append keeps its ink, replace redraws it). The result includes noteUrl, a private link that opens in the user's Wren account after sign-in. Wren emails the user a link to their first note saved here.

Input parameters:

- `context` (object)
- `figure` (object, required)
- `subject` (string): Subject a new note is filed under.
- `target` (object)
- `title` (string): A new note's title.

Output parameters:

- `figureKind` (string)
- `noteId` (string)
- `noteUrl` (string): Private link to the note in the user's Wren account (opens after sign-in).
- `ok` (boolean)
- `pageCount` (number)
- `pageId` (string)
- `pageNumber` (number)
- `title` (string)

### `get_profile` (~37 tokens)

Get Wren profile

Returns the stable Wren account id (Firebase uid) and basic profile of the currently linked account; a returned id confirms the account is linked.

Output parameters:

- `email` (string)
- `id` (string): Stable Wren account id (Firebase uid), not derived from the email address.
- `name` (string)
- `ok` (boolean)
- `signInProvider` (string)
- `uid` (string): Deprecated alias of id.

### `connect_account` (~42 tokens)

Connect your Wren account

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.

Output parameters:

- `linked` (boolean): The Wren account is linked to this ChatGPT account.
- `nextStep` (string): How notes reach the iPad (sign in with the same account).
- `ok` (boolean): true when the account is linked.
- `uid` (string): Linked Wren account id (Firebase uid).

### `list_notes` (~86 tokens)

List Wren notes

Lists the user's Wren notes, most recently edited first, with id, title, subject and page count. The noteId works as target.noteId in write_notebook_page and draw_figure, and as noteId in get_note.

Input parameters:

- `limit` (integer): Max notes to return (default 20).
- `query` (string): Case-insensitive match on title or subject.

Output parameters:

- `notes` (array)
- `ok` (boolean)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/ai-wrenapp-wren/mcp#diagnostics

## Score history

- 2026-11-21: 66
- 2026-11-20: 66
- 2026-11-19: 66
- 2026-11-18: 66
- 2026-11-17: 66
- 2026-11-16: 66
- 2026-11-15: 66
- 2026-11-14: 66

## Common questions

### What is the ai.wrenapp/wren MCP server?

ai.wrenapp/wren is an MCP server listed in the public MCP registry as ai.wrenapp/wren. Ask your AI to write in your Wren notebook: notes, graphs, flowcharts and formulas as real ink. This page covers its hosted endpoint (https://mcp.wrenapp.ai/mcp).

### Is the ai.wrenapp/wren MCP server safe to use?

ai.wrenapp/wren scores 66 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the ai.wrenapp/wren MCP server expose?

ai.wrenapp/wren exposes 8 tools: write_notebook_page, plan_notebook, list_recent_notebooks, get_note, draw_figure, and 3 more. Their descriptions and schemas cost roughly 1,294 tokens of context every time the server is loaded.

### Does the ai.wrenapp/wren MCP server require authentication?

No. We connected to ai.wrenapp/wren without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the ai.wrenapp/wren MCP server still maintained?

ai.wrenapp/wren is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the ai.wrenapp/wren MCP server?

VerifyMCP ran 14 checks on ai.wrenapp/wren, covering TLS certificate, Authorization, HTTPS, HSTS header and 10 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://mcp.wrenapp.ai/mcp
- Website: https://www.wrenapp.ai/
- Changelog RSS feed: https://verifymcp.io/servers/ai-wrenapp-wren/mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-wrenapp-wren/mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/ai-wrenapp-wren/mcp.json
- HTML version of this page: https://verifymcp.io/servers/ai-wrenapp-wren/mcp
