{
  "fixes": {
    "slug": "teable",
    "name": "Teable",
    "listing": "https://www.anchorterminal.com/tools/teable",
    "markdown": "# Fix list: Teable\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/teable, the October 2026 research run, assessed 8 October 2026. Grade C, 61 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 Teable: 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. Payments \u0026 pricing, 30 out of 100, up to 8.8 more on the total\n\nWhy it scored 30: Graded on the hosted cloud. No x402, MPP or L402 found (0). Plan prices are public, Pro at $13 and Business at $20 a seat a month billed yearly, with no per-call price (10). A Free plan with 1,000 rows per space. The pricing page mentions cards only for subscriptions, and we didn't complete a signup (20). Signup is a browser flow, and the security page says Cloudflare Turnstile guards registration, so an agent can't get access by itself (0).\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## 2. Security \u0026 auth, 62 out of 100, up to 6.7 more on the total\n\nWhy it scored 62: OAuth 2.0 with client secret, PKCE and device flows, 10-minute access tokens, rotating refresh tokens and per-action scopes. Personal access tokens need an expiry and take scopes and a list of spaces and projects (30). Read-only tokens are possible per project, and deleted items go to a trash. The API has no approval step for deletes, and the agent skill only instructs a read before each write (14 of 20). Records can hold text written by others and no prompt-injection guidance was found (0 of 15). Record history runs 2 weeks to 3 years by plan and login history 30 days, but the audit log is documented for self-hosted Business plans and above (6 of 15). SECURITY.md sets out private reporting and coordinated disclosure, and one advisory with a CVE was published on 26 August 2026 after its fix. ISO 27001 is the vendor's statement with no certificate linked. No security.txt or bug bounty (12 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## 3. Agent ergonomics, 65 out of 100, up to 5.7 more on the total\n\nWhy it scored 65: Record reads take `projection` to choose fields, `take` to cap rows and `cellFormat=text` for plain values (20 of 25). Paging by `take` and `skip` or a keyset `cursor`, with `filter`, `search`, `orderBy` and `viewId` (20). Errors are JSON with `message`, `status` and a `code` such as `restricted_resource`, and the docs tie 403 to a missing scope. No list of codes is published (12 of 20). No idempotency key or upsert call was found for records. Create, update and delete accept several records in one call, which cuts the writes to retry (5 of 20). Few required parameters and name-keyed fields by default. The only client library is `@teable/openapi` for JavaScript, with an official CLI beside it (8 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. Reliability, 72 out of 100, up to 5.6 more on the total\n\nWhy it scored 72: Graded on Teable Cloud's REST API, with the hosted lines. Status page at status.teable.ai with two monitors and a 90-day view, set up on 28 July 2026, so about 72 days of history exist (20). Its feed lists three incidents. On 27 August 2026 the app and API returned 503s for 5 minutes, on 27 and 28 August the AI gateway was slow for 98 minutes, and on 25 September a maintenance window on AI functions ran 22 minutes over. Only the first touched the API, and all are minor (20). The pricing page and docs give 10 API requests a second on every plan (15). The API answers 429 over the limit and the docs suggest caching and backoff, but Retry-After is documented only for the notifications endpoint and no idempotency key was found (7 of 15). No SLA found (0). The API is generally available with no beta label (10).\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## 5. Schema \u0026 documentation, 75 out of 100, up to 4.1 more on the total\n\nWhy it scored 75: Public OpenAPI 3.0.0 file with 712 paths and 841 operations (25). llms.txt, llms-full.txt and a Markdown copy of every help page (10). All but one operation has a description, and each names the token scopes it needs. They are one or two sentences and don't say when not to use a call (13 of 20). Enums and required fields are used widely, but `filter` and `orderBy` are JSON passed as strings and record `fields` is an untyped object (9 of 15). The four record pages have curl, JavaScript, TypeScript and Python examples. The OpenAPI file documents only 200 and 201 responses apart from one 400, and the error page lists HTTP statuses without a code table (8 of 15). The file's version is fixed at 1.0.0 with no version in the path. A dated public changelog, updated most days, records API behaviour changes (10 of 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## 6. Transparency \u0026 trust, 64 out of 100, up to 3.2 more on the total\n\nMade of editorial 58, provenance 70.\n\nWhy it scored 64: The backend and web app are AGPL-3.0 with added brand terms and the packages are MIT. Enterprise functions and the CLI come from a private repository, and the public one is synced from it (24 of 30). The privacy policy is dated 1 October 2026, gives 30 days for trash, server logs and backups, and names OpenAI and Vercel as AI processors. No DPA was found, and the terms define their subject as the website (18 of 30). No deprecation policy found. The 100-record cap came with a date and kept older keys at 1,000, while other API changes appear in the changelog on the day they ship (6 of 20). Hosting on AWS in Oregon is stated and two AI processors are named, with no full sub-processor list (10 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- Domain age: teable.ai, registered 2025-03-25 (1 year) (3 of 15)\n- Terms of service: read, states 6 of the 7 things a reader expects, and has 3 clauses that cost points (3.1 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, 76 out of 100, up to 2.1 more on the total\n\nWhy it scored 76: The changelog's latest entry is 6 October 2026, 2 days before the check, and the public repository's newest release tag is 2 October (30). The changelog has entries on most days of the last 90 (20). The repository shows 88 open issues, the most recent updated on 1 October 2026, and a community forum exists. We didn't read reply times (12 of 25). `@teable/cli` 0.6.45 came out on 25 September 2026 with 23 versions in 90 days, but `@teable/openapi`, the client in the docs' examples, was last published on 15 September 2025 (7 of 15). The repository has unit, integration and lint workflows and a Renovate config. We didn't read the current CI result (7 of 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- 2 September 2026. Record lists moved to cursor pagination and stopped returning totals by default, and the changelog entry for that day tells API clients to request totals or use the count endpoint. No earlier notice was found (-3). https://help.teable.ai/en/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: whether signup on Teable Cloud asks for a card or an email confirmation. We didn't create an account\n- unchecked: reply times on GitHub issues and the current CI result. The GitHub API refused us, and the count of 88 open issues is read from the issues page\n- unchecked: an ISO 27001 certificate. The security page states certification and no certificate or auditor was named on the pages read\n- unchecked: the monthly-billed prices. The pricing page shows $24 and $36 beside the yearly prices of $13 and $20 with a label of 44 per cent saved, and we read them as the monthly prices\n- unchecked: rate limiting in the source. The public repository has no API throttling code for the 10 requests a second, which appears to live in the enterprise edition\n- No SLA, DPA, sub-processor list, security.txt, bug bounty, deprecation policy, idempotency key, MCP server or prompt-injection guidance was found in the reviewed pages\n- The terms of service define their subject as the website at teable.ai and bar automated agents or scripts from sending requests to it, with no stated exception for API use. They are recorded in provenance because they are the only terms published and cover accounts and AI usage\n- Advisory GHSA-h4hg-45pv-pv8j affected self-hosted instances on local storage, was fixed before publication and credited its reporters. No deduction was taken, a judgement call, because the graded surface is the cloud and the advisory says object-storage deployments are not reachable\n- The status page has existed since 28 July 2026, so its history covers about 72 of the last 90 days\n- An unauthenticated request to a record endpoint returned 403 with `restricted_resource`, not 401\n- The lead was right on vendor, URL and interface. The legal entity is Teable AI, Inc., the API base is app.teable.ai, and an official CLI and agent skill exist that the lead didn't mention\n\n## Weaknesses\n\n- The OpenAPI file documents only 200 and 201 responses for nearly every operation, and the error page lists HTTP statuses without a table of error codes\n- No idempotency key or upsert call was found for record writes, and Retry-After is documented only for the notifications endpoint\n- Teable Cloud allows 10 API requests a second on every plan, Free to Business, with no higher tier published\n- Audit logs are documented for self-hosted Business plans and above, and no SLA, DPA or sub-processor list was found\n- `@teable/openapi`, the JavaScript client the docs use in examples, was last published on 15 September 2025\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- Send `Authorization: Bearer \u003ctoken\u003e` to `https://app.teable.ai/api`. A 403 with code `restricted_resource` usually means a missing scope or a project outside the token's list\n- Page records with `take` of 100 at most. Personal keys created on Teable Cloud since 10 September 2026 are capped at 100 a request\n- Ask for totals from `/api/table/{tableId}/aggregation/row-count`. Record lists stopped returning them by default on 2 September 2026\n- Keep to 10 requests a second, back off on 429, and check before retrying a failed write because no idempotency key exists\n- Pass `fieldKeyType=id` so renamed fields don't break writes, and leave `typecast` off unless values need converting\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": "C",
    "score": 61,
    "assessed": "2026-10-08",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 30,
        "maxGain": 8.8,
        "reason": "Graded on the hosted cloud. No x402, MPP or L402 found (0). Plan prices are public, Pro at $13 and Business at $20 a seat a month billed yearly, with no per-call price (10). A Free plan with 1,000 rows per space. The pricing page mentions cards only for subscriptions, and we didn't complete a signup (20). Signup is a browser flow, and the security page says Cloudflare Turnstile guards registration, so an agent can't get access by itself (0).",
        "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": "security",
        "name": "Security \u0026 auth",
        "score": 62,
        "maxGain": 6.7,
        "reason": "OAuth 2.0 with client secret, PKCE and device flows, 10-minute access tokens, rotating refresh tokens and per-action scopes. Personal access tokens need an expiry and take scopes and a list of spaces and projects (30). Read-only tokens are possible per project, and deleted items go to a trash. The API has no approval step for deletes, and the agent skill only instructs a read before each write (14 of 20). Records can hold text written by others and no prompt-injection guidance was found (0 of 15). Record history runs 2 weeks to 3 years by plan and login history 30 days, but the audit log is documented for self-hosted Business plans and above (6 of 15). SECURITY.md sets out private reporting and coordinated disclosure, and one advisory with a CVE was published on 26 August 2026 after its fix. ISO 27001 is the vendor's statement with no certificate linked. No security.txt or bug bounty (12 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": "ergonomics",
        "name": "Agent ergonomics",
        "score": 65,
        "maxGain": 5.7,
        "reason": "Record reads take `projection` to choose fields, `take` to cap rows and `cellFormat=text` for plain values (20 of 25). Paging by `take` and `skip` or a keyset `cursor`, with `filter`, `search`, `orderBy` and `viewId` (20). Errors are JSON with `message`, `status` and a `code` such as `restricted_resource`, and the docs tie 403 to a missing scope. No list of codes is published (12 of 20). No idempotency key or upsert call was found for records. Create, update and delete accept several records in one call, which cuts the writes to retry (5 of 20). Few required parameters and name-keyed fields by default. The only client library is `@teable/openapi` for JavaScript, with an official CLI beside it (8 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": "reliability",
        "name": "Reliability",
        "score": 72,
        "maxGain": 5.6,
        "reason": "Graded on Teable Cloud's REST API, with the hosted lines. Status page at status.teable.ai with two monitors and a 90-day view, set up on 28 July 2026, so about 72 days of history exist (20). Its feed lists three incidents. On 27 August 2026 the app and API returned 503s for 5 minutes, on 27 and 28 August the AI gateway was slow for 98 minutes, and on 25 September a maintenance window on AI functions ran 22 minutes over. Only the first touched the API, and all are minor (20). The pricing page and docs give 10 API requests a second on every plan (15). The API answers 429 over the limit and the docs suggest caching and backoff, but Retry-After is documented only for the notifications endpoint and no idempotency key was found (7 of 15). No SLA found (0). The API is generally available with no beta label (10).",
        "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": "schema",
        "name": "Schema \u0026 documentation",
        "score": 75,
        "maxGain": 4.1,
        "reason": "Public OpenAPI 3.0.0 file with 712 paths and 841 operations (25). llms.txt, llms-full.txt and a Markdown copy of every help page (10). All but one operation has a description, and each names the token scopes it needs. They are one or two sentences and don't say when not to use a call (13 of 20). Enums and required fields are used widely, but `filter` and `orderBy` are JSON passed as strings and record `fields` is an untyped object (9 of 15). The four record pages have curl, JavaScript, TypeScript and Python examples. The OpenAPI file documents only 200 and 201 responses apart from one 400, and the error page lists HTTP statuses without a code table (8 of 15). The file's version is fixed at 1.0.0 with no version in the path. A dated public changelog, updated most days, records API behaviour changes (10 of 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": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 64,
        "maxGain": 3.2,
        "reason": "The backend and web app are AGPL-3.0 with added brand terms and the packages are MIT. Enterprise functions and the CLI come from a private repository, and the public one is synced from it (24 of 30). The privacy policy is dated 1 October 2026, gives 30 days for trash, server logs and backups, and names OpenAI and Vercel as AI processors. No DPA was found, and the terms define their subject as the website (18 of 30). No deprecation policy found. The 100-record cap came with a date and kept older keys at 1,000, while other API changes appear in the changelog on the day they ship (6 of 20). Hosting on AWS in Oregon is stated and two AI processors are named, with no full sub-processor list (10 of 20).",
        "blend": "editorial 58, provenance 70",
        "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": 76,
        "maxGain": 2.1,
        "reason": "The changelog's latest entry is 6 October 2026, 2 days before the check, and the public repository's newest release tag is 2 October (30). The changelog has entries on most days of the last 90 (20). The repository shows 88 open issues, the most recent updated on 1 October 2026, and a community forum exists. We didn't read reply times (12 of 25). `@teable/cli` 0.6.45 came out on 25 September 2026 with 23 versions in 90 days, but `@teable/openapi`, the client in the docs' examples, was last published on 15 September 2025 (7 of 15). The repository has unit, integration and lint workflows and a Renovate config. We didn't read the current CI result (7 of 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": "Domain age",
        "value": "teable.ai, registered 2025-03-25 (1 year)",
        "points": 3,
        "max": 15
      },
      {
        "label": "Terms of service",
        "value": "read, states 6 of the 7 things a reader expects, and has 3 clauses that cost points",
        "points": 3.1,
        "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": [
      "2 September 2026. Record lists moved to cursor pagination and stopped returning totals by default, and the changelog entry for that day tells API clients to request totals or use the count endpoint. No earlier notice was found (-3). https://help.teable.ai/en/changelog"
    ],
    "unchecked": [
      "unchecked: whether signup on Teable Cloud asks for a card or an email confirmation. We didn't create an account",
      "unchecked: reply times on GitHub issues and the current CI result. The GitHub API refused us, and the count of 88 open issues is read from the issues page",
      "unchecked: an ISO 27001 certificate. The security page states certification and no certificate or auditor was named on the pages read",
      "unchecked: the monthly-billed prices. The pricing page shows $24 and $36 beside the yearly prices of $13 and $20 with a label of 44 per cent saved, and we read them as the monthly prices",
      "unchecked: rate limiting in the source. The public repository has no API throttling code for the 10 requests a second, which appears to live in the enterprise edition",
      "No SLA, DPA, sub-processor list, security.txt, bug bounty, deprecation policy, idempotency key, MCP server or prompt-injection guidance was found in the reviewed pages",
      "The terms of service define their subject as the website at teable.ai and bar automated agents or scripts from sending requests to it, with no stated exception for API use. They are recorded in provenance because they are the only terms published and cover accounts and AI usage",
      "Advisory GHSA-h4hg-45pv-pv8j affected self-hosted instances on local storage, was fixed before publication and credited its reporters. No deduction was taken, a judgement call, because the graded surface is the cloud and the advisory says object-storage deployments are not reachable",
      "The status page has existed since 28 July 2026, so its history covers about 72 of the last 90 days",
      "An unauthenticated request to a record endpoint returned 403 with `restricted_resource`, not 401",
      "The lead was right on vendor, URL and interface. The legal entity is Teable AI, Inc., the API base is app.teable.ai, and an official CLI and agent skill exist that the lead didn't mention"
    ],
    "weaknesses": [
      "The OpenAPI file documents only 200 and 201 responses for nearly every operation, and the error page lists HTTP statuses without a table of error codes",
      "No idempotency key or upsert call was found for record writes, and Retry-After is documented only for the notifications endpoint",
      "Teable Cloud allows 10 API requests a second on every plan, Free to Business, with no higher tier published",
      "Audit logs are documented for self-hosted Business plans and above, and no SLA, DPA or sub-processor list was found",
      "`@teable/openapi`, the JavaScript client the docs use in examples, was last published on 15 September 2025"
    ],
    "agentNotes": [
      "Send `Authorization: Bearer \u003ctoken\u003e` to `https://app.teable.ai/api`. A 403 with code `restricted_resource` usually means a missing scope or a project outside the token's list",
      "Page records with `take` of 100 at most. Personal keys created on Teable Cloud since 10 September 2026 are capped at 100 a request",
      "Ask for totals from `/api/table/{tableId}/aggregation/row-count`. Record lists stopped returning them by default on 2 September 2026",
      "Keep to 10 requests a second, back off on 429, and check before retrying a failed write because no idempotency key exists",
      "Pass `fieldKeyType=id` so renamed fields don't break writes, and leave `typecast` off unless values need converting"
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-08",
    "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"
  }
}