Excalidraw

# Fix list: Excalidraw

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/excalidraw, the October 2026 research run, assessed 9 October 2026. Grade C, 54.5 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Excalidraw: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Reliability, 50 out of 100, up to 10 more on the total

Why it scored 50: Scored on the hosted lines for the Excalidraw+ API and MCP server. A status page exists at status.excalidraw.com (20). Its robots.txt disallows every path, so the incident history was not read (5). The rate limit is 600 requests a minute per IP, with `X-RateLimit-Limit`, `X-RateLimit-Remaining` and `X-RateLimit-Reset` headers (15). The docs ask for backoff on 429 and show a retry that waits for the reset time. No idempotency keys were found, `PUT` content replacement is repeatable and `PATCH` merges by element version (10). No SLA found (0). Both surfaces are in public beta (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 2. Payments & pricing, 35 out of 100, up to 8.1 more on the total

Why it scored 35: No x402, MPP or L402 (0). Plan prices are public, Free at $0 and Plus at $6 a user a month, with no per-call price (10). The free editor and the keyless MCP App need no card, while the API needs the Plus plan, whose 14-day trial does not say whether a card is taken (15). An agent can call the free MCP App at https://mcp.excalidraw.com with no signup, but an API key is created by a person in workspace settings (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 3. Agent ergonomics, 58 out of 100, up to 6.8 more on the total

Why it scored 58: The docs list 37 MCP tools (5). The server shows a key only the tools its routes allow, so a read-only or route-restricted key cuts the list (7). Lists take `offset` and `limit` up to 100, logs take a cursor and filters, and `search_scene_content` returns only matching elements (18). Errors are JSON with `statusCode`, `error` and `message`, and seven status codes are documented (13). No idempotency keys. Deletes of scenes and collections move them to trash, and MCP annotations could not be read (8). `create_diagram` measures labels and routes arrows, but there is no SDK for the API (7).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 4. Schema & documentation, 65 out of 100, up to 5.7 more on the total

Why it scored 65: The endpoint reference is generated from an OpenAPI schema, but no file is linked from the docs, and the MCP tool definitions could not be read without a key (10). `llms.txt` and `llms-full.txt` are published (10). The tools page says when to use `search_scene_content` over `get_scene_content`, `label` over `text` and `create_diagram` over `edit_scene_content` (16). Parameters carry ranges and enums in the reference, while `edit_scene_content` takes its elements as a JSON array string (9). Each endpoint has samples in six languages and error bodies for 400, 401 and 403 (12). The path is versioned as `/api/v1` and a monthly product changelog carries API and MCP items, with no API changelog of its own (8).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 5. Maintenance & community, 44 out of 100, up to 4.9 more on the total

Why it scored 44: The newest changelog entry is dated 1 September 2026, 38 days before the check (20). Two dated entries fall in the last 90 days, 31 July and 1 September (0). A public changelog, a roadmap, a Discord server and support@excalidraw.com exist, and the editor repository had commits on 8 October 2026 with 3,251 open issues and pull requests (12). No vendor entry was found in the official MCP registry and there is no SDK for the API. The editor package `@excalidraw/excalidraw` 0.18.1 dates from 20 April 2026 (5). The editor repository runs test, lint and size workflows, and the MCP App repository has no CI and no commit since 24 March 2026 (7).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## 6. Security & auth, 73 out of 100, up to 4.7 more on the total

Why it scored 73: API keys are workspace-bound, take `read` or `full` permission or specific routes, carry an expiry and can be rotated or soft-deleted. Personal and workspace keys differ in access to private collections (30). Read-only keys expose only read tools and deletes go to trash, with no confirmation step found (14). Scene text written by collaborators returns to the model and `add_image` fetches a public URL, with no injection guidance found (3). `GET /logs` returns the workspace audit log with filters by user, action and operation (13). Excalidraw states SOC 2 Type 1 and Type 2, yearly penetration tests and Vanta scanning, and publishes advisories on GitHub. No security.txt, bug bounty or SECURITY.md was found, and the trust centre report was not read (13).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 7. Transparency & trust, 63 out of 100, up to 3.2 more on the total

Made of editorial 45, provenance 80.

Why it scored 63: The graded service is closed under public terms of use, while the editor and the MCP App are MIT (20). The privacy policy of 29 April 2021 says servers are in the United States and data is kept at most one month after an account ends. The DPA of 13 January 2026 says cancelling a subscription does not delete data and retention follows internal policies. Neither names AI providers (14). No deprecation policy, only the beta warning that contracts may break (3). The DPA points to a sub-processor list at trust.excalidraw.com, which is script-drawn and was not read, and says some sub-processors are in the United States (8).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: excalidraw.com, registered 2020-01-03 (6 years) (11 of 15)
- Terms of service: read, states 5 of the 7 things a reader expects, and has 2 clauses that cost points (4.3 of 10)
- security.txt: not found (0 of 10)

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- Checked 2026-10-09. GHSA-39h7-pwv7-rc3x (CVE-2025-54881, medium), cross-site scripting in the Mermaid to Excalidraw conversion step, was published on 21 April 2026 and fixed in `@excalidraw/excalidraw` 0.18.1 and `@excalidraw/mermaid-to-excalidraw` 1.1.3. Fixed and published, so 2 points (https://github.com/excalidraw/excalidraw/security/advisories/GHSA-39h7-pwv7-rc3x).

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- unchecked: the incident history on status.excalidraw.com. Its robots.txt disallows every path, so Reliability takes 5 of 30 for the record
- unchecked: the MCP tool definitions, annotations and their size on https://api.excalidraw.com/api/v1/mcp. They need a key, and the host's robots.txt disallows every path, so no request was sent
- unchecked: the sub-processor list and SOC 2 report in the trust centre at trust.excalidraw.com, which is script-drawn
- unchecked: any page of excalidraw.com beyond its robots.txt, which carries both `Allow: /` and `Disallow: /` for unnamed agents. Its security.txt was not requested
- Whether the 14-day Plus trial takes a card. The terms say the account is charged when the trial ends
- Whether the OpenAPI schema behind the reference pages is published at a public address. No link was found in the docs or `llms.txt`
- The terms of use forbid access through a bot or script, and the robots.txt files of api.excalidraw.com and status.excalidraw.com disallow every path. Both matter before any probe is run
- The lead named only the free MCP App and the npm package. The Excalidraw+ REST API and MCP server, in public beta since the May 2026 changelog entry, are the documented agent surface and are what the scores cover

## Weaknesses

- The API and MCP server are in public beta. The docs say endpoints, tool names and schemas may change and to expect breaking changes
- API and MCP access needs the Plus plan at $6 a user a month. No per-call price, and no SDK for the API
- No OpenAPI file is linked from the docs, although the reference pages are generated from one. No idempotency keys and no SLA found
- The terms of use (in effect 1 June 2022) forbid access through a bot or script and do not mention the API. This matters before any probe is run
- The privacy policy is dated 29 April 2021 and names no AI providers. The sub-processor list sits in a script-drawn trust centre
- In the free MCP App, `create_view` is marked read-only yet stores a checkpoint for 30 days, and its reply names a `read_widget_context` tool the server does not register

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Send `Authorization: Bearer <API_KEY>` to https://api.excalidraw.com/api/v1. For MCP, use `POST /api/v1/mcp` only. The server is stateless and answers 405 to other methods
- Use a personal key to reach the owner's private collection with the collection ID `private`. A workspace admin must enable personal keys first
- Call `read_diagram_format`, `read_presentation_format` or `read_freeform_format` before the first scene write in a session
- Write with `edit_scene_content` and bind arrows through `startBinding` and `endBinding` with `tempId` references. `PUT /scenes/{sceneId}/content` removes every element left out of the request
- Stay under 600 requests a minute per IP and wait until `X-RateLimit-Reset` after a 429

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.