{
  "fixes": {
    "slug": "mattermost",
    "name": "Mattermost",
    "listing": "https://www.anchorterminal.com/tools/mattermost",
    "markdown": "# Fix list: Mattermost\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/mattermost, the October 2026 research run, assessed 9 October 2026. Grade B, 65.8 out of 100.\n\nThis 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.\n\nFor a coding agent working on Mattermost: 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.\n\n## 1. Security \u0026 auth, 63 out of 100, up to 6.5 more on the total\n\nWhy it scored 63: Personal access tokens are named, several per account, and can be revoked or disabled by the user or an admin. They carry no scopes and never expire, and a token for a System Admin account has full admin rights. OAuth 2.0 with PKCE and optional dynamic client registration has no scopes either. Tokens travel in the `Authorization` header or a cookie, with no query-string option in the docs (20 of 30). Bot accounts are separate accounts with a role, creating them and creating tokens are both off by default, and roles and permission schemes bound what an account can do. No read-only token and no confirmation step for deletes on the REST API (13 of 20). Messages are other people's text and no prompt-injection guidance was found in the API or Agents plugin docs (2 of 15). Audit logging records REST API and mmctl activity to a JSON schema. The docs label it Beta and it is not written to a file by default (10 of 15). A disclosure policy, a public Bugcrowd bounty, annual penetration tests, and a SOC 2 Type II report (2025) and ISO 27001:2022 certificate listed on the trust centre. No security.txt (18 of 20).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-security):\n\n- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.\n\nModels 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.\n\n## 2. Payments \u0026 pricing, 50 out of 100, up to 6.3 more on the total\n\nWhy it scored 50: Read with the self-hosted rule. No x402, MPP or L402 (0). Team Edition and the Entry edition are free and their limits are published. Professional, Enterprise and Enterprise Advanced show Contact Sales, Get Pricing and Request Quote, with no price, so half marks (10). Free to run with no card and no vendor account (20). An owner can start the Docker image, and the docs say `POST /api/v4/users` needs no permission on an open server and that the first account becomes System Admin, so no browser signup with the vendor is needed (20).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):\n\nThe published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).\n\n- 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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).\n\nPayment 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.\n\nOpen-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.\n\n## 3. Agent ergonomics, 63 out of 100, up to 6 more on the total\n\nWhy it scored 63: List endpoints take `per_page` up to 200 with a default of 60, and channel posts take `since`, `before` and `after`. No field selection, and posts come back whole with metadata (14 of 25). 103 `page` or `per_page` parameters across the source, plus search endpoints for posts, users, channels and files (17 of 20). Errors are JSON with `id`, `message`, `request_id` and `status_code`. The ids are stable strings with no published list, and the 429 body is plain text (14 of 20). No idempotency key was found. The post schema has an undescribed `pending_post_id`, and nothing documents safe retries for writes (4 of 20). A post needs only `channel_id` and `message`, `me` stands in for the caller's user id, and the official drivers are TypeScript (`@mattermost/client`) and Go (14 of 15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):\n\n- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.\n\nModels 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.\n\n## 4. Schema \u0026 documentation, 76 out of 100, up to 3.9 more on the total\n\nWhy it scored 76: The OpenAPI 3.0 source is public in the repository, 57 YAML path files with 604 operations on 485 paths on the master branch. The reference site counts 549 endpoints for the released version (25). No llms.txt on mattermost.com, docs.mattermost.com or developers.mattermost.com. The docs are MDX files in the public repository and are served as HTML (2 of 10). Every operation has a description, median 192 characters. 548 state the permission needed and 304 the minimum server version, and few say when not to use an endpoint (15 of 20). Bodies are JSON with 848 required lists, but only 31 enums and 29 length or range limits across the source (9 of 15). The reference says each endpoint page carries samples in curl, PowerShell, Python, Node and Go, and a quick-start page has six curl recipes. Most operations list 400, 401 and 403 with one shared error shape, and error ids are not catalogued (10 of 15). The API is at v4, and the server changelog is dated with an API Changes list per release (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):\n\nAPIs and MCP servers.\n\n- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.\n\nModels 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.\n\n## 5. Reliability, 83 out of 100, up to 3.4 more on the total\n\nWhy it scored 83: Read with the local-software lines, since the graded surface is the server its owner hosts. Mattermost Cloud is sold through sales only. Official tarballs, Docker images and a Kubernetes operator, with Ubuntu LTS 22.04 or later and PostgreSQL 14 or later stated (20). Public workflows run server, web app, end-to-end and mmctl tests on every push to master, with CodeQL and Scorecards. We did not read the result of the latest run (20 of 25). 612 issues were open on 9 October 2026. The 25 newest were opened between 5 September and 8 October, one of them a server crash when disabling a bot that owns itself, and reply counts were not read (16 of 25). Feature releases come monthly and each changelog entry has a Breaking Changes block and an API Changes list, but removals land in minor versions, such as `POST /api/v4/posts/ids/reactions` in 11.11 (12 of 15). Version 11.11, with API v4 described as stable (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):\n\nHosted APIs, MCP servers, models and platforms.\n\n- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 10, the surface agents use is generally available, not beta or preview.\n\nLocal packages, SDKs, frameworks and stdio MCP servers.\n\n- 20, installs from an official package with supported runtimes stated.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 15, version 1.0 or later, or declared stable.\n\nProtocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.\n\n## 6. Transparency \u0026 trust, 75 out of 100, up to 2.2 more on the total\n\nMade of editorial 80, provenance 69.\n\nWhy it scored 75: Team Edition is MIT as a compiled binary, the source is AGPL v3 or a commercial licence with parts under Apache 2.0, and the Enterprise Edition binary that Entry runs is under a commercial licence. Open core, with the split stated in `LICENSE.txt` (26 of 30). On a self-hosted server the data stays with the owner. The privacy policy of 30 December 2024 is general and gives no retention period in numbers, the Software and Services Licence Agreement of 12 January 2023 covers cloud and on-premise use, and the DPA is a linked document we did not read (19 of 30). A Removed and Deprecated Features page lists upcoming removals by version and month, and each release has a published support end date, with extended support releases kept for 12 months (18 of 20). The telemetry page lists what a server sends, says it is on by default and gives the System Console switches to turn it off on a self-hosted server. Cloud admins cannot turn it off (17 of 20).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):\n\n- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).\n\nThe other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.\n\nProvenance checks not met in full (half of this category, computed from checked facts):\n\n- Endpoint on the vendor's domain:  is not on mattermost.com (0 of 15)\n- Terms of service: read, states 4 of the 7 things a reader expects, and has 1 clause that costs points (5.4 of 10)\n- Privacy policy: read, states 6 of the 8 things a reader expects (8.5 of 10)\n- security.txt: not found (0 of 10)\n\n## 7. Maintenance \u0026 community, 89 out of 100, up to 1 more on the total\n\nWhy it scored 89: Server 11.7.12 was released on 8 October 2026, one day before the check, and 11.11.1 on 24 September (30). Sixteen releases since 11 July 2026 by the dated changelog (20). 612 open issues on a repository with 39,301 stars. Several of the newest 25 were updated after they were opened, and a forum and a community server are linked. We did not read reply counts (17 of 25). Official TypeScript and Go drivers. `@mattermost/client` on npm is at 11.9.0 while the server is at 11.11, and the Go module is tagged `server/public/v0.4.4` (12 of 15). CI, CodeQL and Scorecards workflows with actions pinned by commit, and an SBOM published for each release (10).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):\n\n- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.\n\nModels are read for deprecation notice periods and model churn rather than release counts.\n\n## Deductions\n\nEach comes off the total. A fixed and documented problem counts for less at the next check.\n\n- 2025-10-28 to 2026-10-08. The v11 changelog marks 52 dot releases in twelve months as carrying security fixes. Three are rated critical, 11.0.4 on 28 October 2025 in the server and 11.0.6 and 11.1.1 on 21 November 2025 in the bundled Jira plugin, and 21 more include high-severity fixes. All were fixed in a release and announced, with details published 30 days later, so 5 points. The advisory table itself is drawn by script and was not read (https://docs.mattermost.com/product-overview/mattermost-v11-changelog).\n\n## What we couldn't check\n\nWhat 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.\n\n- unchecked: the rows of the security updates table (issue ids, CVE numbers and severities per advisory). The page draws them by script from a feed we did not fetch, so the count comes from the changelog's per-release notes\n- unchecked: whether the latest CI run on the master branch passed. The workflows were read, the run results were not\n- unchecked: reply counts and response times on GitHub issues\n- unchecked: the Data Processing Addendum, the SOC 2 report and the other trust centre documents, which sit behind an access request or a link we did not follow\n- unchecked: the publish date of `@mattermost/client` 11.9.0 on npm, and whether the MCP server has an entry in the official MCP registry\n- unchecked: the terms the API introduction links at about.mattermost.com/default-terms, which were not requested\n- The lead named a self-hosted option beside a hosted one. On 9 October 2026 the pricing page shows no price for any paid edition and its FAQ says Mattermost Cloud Enterprise is single-tenant and sold through sales, so the grade is on the self-hosted server\n- The lead's docs link, docs.mattermost.com/api, is right. api.mattermost.com redirects to it\n- The 99.9 per cent uptime SLA in the docs is for the Cloud dedicated add-on only, and status.mattermost.com covers the vendor's cloud and sites, not a server its owner runs\n- The MCP server ships inside the Agents plugin. Its tool definitions were read in the plugin's docs and readme only, and the plugin was not installed or run. Write tools need an Enterprise licence, which Entry counts as\n- Whether the free Entry edition's limits marked as coming in a future release (board cards, playbook runs, agent queries, call length, push notifications) are enforced yet was not established\n- No prompt-injection guidance, llms.txt or security.txt was found on the pages and files read\n\n## Weaknesses\n\n- Personal access tokens have no scopes and no expiry. A token does whatever its account can do\n- 52 dot releases from 28 October 2025 to 8 October 2026 carried security fixes, three rated critical and 21 including high-severity fixes\n- No public price for Professional, Enterprise or Enterprise Advanced, and Mattermost Cloud is sold through sales only\n- No idempotency key on `POST /api/v4/posts`, so a retried send can post twice\n- Rate limiting is off by default on a self-hosted server, and no llms.txt or security.txt was found\n\n## What costs an agent a turn today\n\nThe notes we give agents before they call it. Each one is a workaround an agent shouldn't need.\n\n- Ask the system admin for a bot account and its token. Bot creation and personal access tokens are both off until enabled in the System Console\n- Send `Authorization: Bearer \u003ctoken\u003e` to `https://\u003cserver\u003e/api/v4`. Use `me` in place of a user id for the token's own account\n- Page with `page` and `per_page`. The maximum is 200, the default 60, and larger values are cut without an error\n- Read the error `id` and `status_code`. A 501 means the server's edition or licence does not include that endpoint\n- If the server has rate limiting on, read `X-Ratelimit-Remaining` and `X-Ratelimit-Reset`. The 429 body is the plain text `limit exceeded`\n\n## When it's done\n\nSend 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.\n",
    "grade": "B",
    "score": 65.8,
    "assessed": "2026-10-09",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "security",
        "name": "Security \u0026 auth",
        "score": 63,
        "maxGain": 6.5,
        "reason": "Personal access tokens are named, several per account, and can be revoked or disabled by the user or an admin. They carry no scopes and never expire, and a token for a System Admin account has full admin rights. OAuth 2.0 with PKCE and optional dynamic client registration has no scopes either. Tokens travel in the `Authorization` header or a cookie, with no query-string option in the docs (20 of 30). Bot accounts are separate accounts with a role, creating them and creating tokens are both off by default, and roles and permission schemes bound what an account can do. No read-only token and no confirmation step for deletes on the REST API (13 of 20). Messages are other people's text and no prompt-injection guidance was found in the API or Agents plugin docs (2 of 15). Audit logging records REST API and mmctl activity to a JSON schema. The docs label it Beta and it is not written to a file by default (10 of 15). A disclosure policy, a public Bugcrowd bounty, annual penetration tests, and a SOC 2 Type II report (2025) and ISO 27001:2022 certificate listed on the trust centre. No security.txt (18 of 20).",
        "checklist": [
          "- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-security"
      },
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 50,
        "maxGain": 6.3,
        "reason": "Read with the self-hosted rule. No x402, MPP or L402 (0). Team Edition and the Entry edition are free and their limits are published. Professional, Enterprise and Enterprise Advanced show Contact Sales, Get Pricing and Request Quote, with no price, so half marks (10). Free to run with no card and no vendor account (20). An owner can start the Docker image, and the docs say `POST /api/v4/users` needs no permission on an open server and that the first account becomes System Admin, so no browser signup with the vendor is needed (20).",
        "checklist": [
          "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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-payments"
      },
      {
        "key": "ergonomics",
        "name": "Agent ergonomics",
        "score": 63,
        "maxGain": 6,
        "reason": "List endpoints take `per_page` up to 200 with a default of 60, and channel posts take `since`, `before` and `after`. No field selection, and posts come back whole with metadata (14 of 25). 103 `page` or `per_page` parameters across the source, plus search endpoints for posts, users, channels and files (17 of 20). Errors are JSON with `id`, `message`, `request_id` and `status_code`. The ids are stable strings with no published list, and the 429 body is plain text (14 of 20). No idempotency key was found. The post schema has an undescribed `pending_post_id`, and nothing documents safe retries for writes (4 of 20). A post needs only `channel_id` and `message`, `me` stands in for the caller's user id, and the official drivers are TypeScript (`@mattermost/client`) and Go (14 of 15).",
        "checklist": [
          "- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-ergonomics"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 76,
        "maxGain": 3.9,
        "reason": "The OpenAPI 3.0 source is public in the repository, 57 YAML path files with 604 operations on 485 paths on the master branch. The reference site counts 549 endpoints for the released version (25). No llms.txt on mattermost.com, docs.mattermost.com or developers.mattermost.com. The docs are MDX files in the public repository and are served as HTML (2 of 10). Every operation has a description, median 192 characters. 548 state the permission needed and 304 the minimum server version, and few say when not to use an endpoint (15 of 20). Bodies are JSON with 848 required lists, but only 31 enums and 29 length or range limits across the source (9 of 15). The reference says each endpoint page carries samples in curl, PowerShell, Python, Node and Go, and a quick-start page has six curl recipes. Most operations list 400, 401 and 403 with one shared error shape, and error ids are not catalogued (10 of 15). The API is at v4, and the server changelog is dated with an API Changes list per release (15).",
        "checklist": [
          "APIs and MCP servers.",
          "- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-schema"
      },
      {
        "key": "reliability",
        "name": "Reliability",
        "score": 83,
        "maxGain": 3.4,
        "reason": "Read with the local-software lines, since the graded surface is the server its owner hosts. Mattermost Cloud is sold through sales only. Official tarballs, Docker images and a Kubernetes operator, with Ubuntu LTS 22.04 or later and PostgreSQL 14 or later stated (20). Public workflows run server, web app, end-to-end and mmctl tests on every push to master, with CodeQL and Scorecards. We did not read the result of the latest run (20 of 25). 612 issues were open on 9 October 2026. The 25 newest were opened between 5 September and 8 October, one of them a server crash when disabling a bot that owns itself, and reply counts were not read (16 of 25). Feature releases come monthly and each changelog entry has a Breaking Changes block and an API Changes list, but removals land in minor versions, such as `POST /api/v4/posts/ids/reactions` in 11.11 (12 of 15). Version 11.11, with API v4 described as stable (15).",
        "checklist": [
          "Hosted APIs, MCP servers, models and platforms.",
          "- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 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.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-reliability"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 75,
        "maxGain": 2.2,
        "reason": "Team Edition is MIT as a compiled binary, the source is AGPL v3 or a commercial licence with parts under Apache 2.0, and the Enterprise Edition binary that Entry runs is under a commercial licence. Open core, with the split stated in `LICENSE.txt` (26 of 30). On a self-hosted server the data stays with the owner. The privacy policy of 30 December 2024 is general and gives no retention period in numbers, the Software and Services Licence Agreement of 12 January 2023 covers cloud and on-premise use, and the DPA is a linked document we did not read (19 of 30). A Removed and Deprecated Features page lists upcoming removals by version and month, and each release has a published support end date, with extended support releases kept for 12 months (18 of 20). The telemetry page lists what a server sends, says it is on by default and gives the System Console switches to turn it off on a self-hosted server. Cloud admins cannot turn it off (17 of 20).",
        "blend": "editorial 80, provenance 69",
        "checklist": [
          "- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-transparency"
      },
      {
        "key": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 89,
        "maxGain": 1,
        "reason": "Server 11.7.12 was released on 8 October 2026, one day before the check, and 11.11.1 on 24 September (30). Sixteen releases since 11 July 2026 by the dated changelog (20). 612 open issues on a repository with 39,301 stars. Several of the newest 25 were updated after they were opened, and a forum and a community server are linked. We did not read reply counts (17 of 25). Official TypeScript and Go drivers. `@mattermost/client` on npm is at 11.9.0 while the server is at 11.11, and the Go module is tagged `server/public/v0.4.4` (12 of 15). CI, CodeQL and Scorecards workflows with actions pinned by commit, and an SBOM published for each release (10).",
        "checklist": [
          "- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.",
          "Models are read for deprecation notice periods and model churn rather than release counts."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-maintenance"
      }
    ],
    "provenance": [
      {
        "label": "Endpoint on the vendor's domain",
        "value": " is not on mattermost.com",
        "points": 0,
        "max": 15
      },
      {
        "label": "Terms of service",
        "value": "read, states 4 of the 7 things a reader expects, and has 1 clause that costs points",
        "points": 5.4,
        "max": 10
      },
      {
        "label": "Privacy policy",
        "value": "read, states 6 of the 8 things a reader expects",
        "points": 8.5,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "not found",
        "points": 0,
        "max": 10
      }
    ],
    "deductions": [
      "2025-10-28 to 2026-10-08. The v11 changelog marks 52 dot releases in twelve months as carrying security fixes. Three are rated critical, 11.0.4 on 28 October 2025 in the server and 11.0.6 and 11.1.1 on 21 November 2025 in the bundled Jira plugin, and 21 more include high-severity fixes. All were fixed in a release and announced, with details published 30 days later, so 5 points. The advisory table itself is drawn by script and was not read (https://docs.mattermost.com/product-overview/mattermost-v11-changelog)."
    ],
    "unchecked": [
      "unchecked: the rows of the security updates table (issue ids, CVE numbers and severities per advisory). The page draws them by script from a feed we did not fetch, so the count comes from the changelog's per-release notes",
      "unchecked: whether the latest CI run on the master branch passed. The workflows were read, the run results were not",
      "unchecked: reply counts and response times on GitHub issues",
      "unchecked: the Data Processing Addendum, the SOC 2 report and the other trust centre documents, which sit behind an access request or a link we did not follow",
      "unchecked: the publish date of `@mattermost/client` 11.9.0 on npm, and whether the MCP server has an entry in the official MCP registry",
      "unchecked: the terms the API introduction links at about.mattermost.com/default-terms, which were not requested",
      "The lead named a self-hosted option beside a hosted one. On 9 October 2026 the pricing page shows no price for any paid edition and its FAQ says Mattermost Cloud Enterprise is single-tenant and sold through sales, so the grade is on the self-hosted server",
      "The lead's docs link, docs.mattermost.com/api, is right. api.mattermost.com redirects to it",
      "The 99.9 per cent uptime SLA in the docs is for the Cloud dedicated add-on only, and status.mattermost.com covers the vendor's cloud and sites, not a server its owner runs",
      "The MCP server ships inside the Agents plugin. Its tool definitions were read in the plugin's docs and readme only, and the plugin was not installed or run. Write tools need an Enterprise licence, which Entry counts as",
      "Whether the free Entry edition's limits marked as coming in a future release (board cards, playbook runs, agent queries, call length, push notifications) are enforced yet was not established",
      "No prompt-injection guidance, llms.txt or security.txt was found on the pages and files read"
    ],
    "weaknesses": [
      "Personal access tokens have no scopes and no expiry. A token does whatever its account can do",
      "52 dot releases from 28 October 2025 to 8 October 2026 carried security fixes, three rated critical and 21 including high-severity fixes",
      "No public price for Professional, Enterprise or Enterprise Advanced, and Mattermost Cloud is sold through sales only",
      "No idempotency key on `POST /api/v4/posts`, so a retried send can post twice",
      "Rate limiting is off by default on a self-hosted server, and no llms.txt or security.txt was found"
    ],
    "agentNotes": [
      "Ask the system admin for a bot account and its token. Bot creation and personal access tokens are both off until enabled in the System Console",
      "Send `Authorization: Bearer \u003ctoken\u003e` to `https://\u003cserver\u003e/api/v4`. Use `me` in place of a user id for the token's own account",
      "Page with `page` and `per_page`. The maximum is 200, the default 60, and larger values are cut without an error",
      "Read the error `id` and `status_code`. A 501 means the server's edition or licence does not include that endpoint",
      "If the server has rate limiting on, read `X-Ratelimit-Remaining` and `X-Ratelimit-Reset`. The 429 body is the plain text `limit exceeded`"
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-10",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.4",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  }
}