{
  "fixes": {
    "slug": "permit-mcp-gateway",
    "name": "Permit MCP Gateway",
    "listing": "https://www.anchorterminal.com/tools/permit-mcp-gateway",
    "markdown": "# Fix list: Permit MCP Gateway\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/permit-mcp-gateway, the October 2026 research run, assessed 1 October 2026. Grade C, 54.5 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 Permit MCP Gateway: 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, 10 out of 100, up to 11.3 more on the total\n\nWhy it scored 10: No machine payment protocol (0). Approvals are Enterprise only through a demo, and the gateway has no public price (0). Evaluation starts free on the hosted gateway, but approvals aren't in it, so 10 of 20. A person signs in through the browser consent flow (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. Reliability, 47 out of 100, up to 10.6 more on the total\n\nWhy it scored 47: Permit's status page at permit-io.instatus.com lists the backend, OPAL, frontend, website and PDP services, at 100 per cent, but not the gateway or agent.security (10 of 20). With no gateway component there's no incident history for it (5). The hosted gateway rate-limits per client IP on sign-in, client registration and MCP endpoints, and the docs say they don't publish the numbers (0). A 429 carries a `rate_limited` JSON body and the docs say to back off and retry with growing waits (12 of 15). Enterprise has a 99.99 per cent uptime option (10). Approvals carry an Enterprise label, not beta (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## 3. Schema \u0026 documentation, 53 out of 100, up to 7.6 more on the total\n\nWhy it scored 53: The gateway adds no tools of its own and passes upstream tool schemas through. Permit's management API has an OpenAPI-backed reference (15 of 25). No llms.txt or Markdown copies of the docs found (0). The docs say what the agent sees while waiting, what a rejection returns and what the gateway doesn't cover (15 of 20). Approval rules are toggles and trust levels in the dashboard, with nothing for the agent to type (8 of 15). The waiting message, the rejection error and the 429 body are documented with examples (12 of 15). The product changelog on Canny has no entry after May 2024, and the gateway has no changelog (3 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## 4. Maintenance \u0026 community, 43 out of 100, up to 5 more on the total\n\nWhy it scored 43: No release notes for the gateway. The docs record new gateway capability on 2026-07-28 and 2026-07-30 (the HTTP egress proxy) and a rewrite on 17 and 20 September (20 of 30). Those are docs commits, not releases or changelog entries, so part credit for visible activity (5 of 20). Closed service with a Slack community and support email, and the public changelog stopped in May 2024 (5 of 15). Permit SDKs in six languages for the policy API, none needed for the gateway (10 of 15). Nothing to install, so we score package health on the public docs repository's CI alone (3 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## 5. Transparency \u0026 trust, 45 out of 100, up to 4.8 more on the total\n\nMade of editorial 40, provenance 50.\n\nWhy it scored 45: Closed service under terms updated 2026-07-01 that name Permit Inc., a Delaware corporation, under Delaware law (15 of 30). The docs say hosted traffic, including tool arguments and upstream responses, passes through Permit's infrastructure, and that customer-controlled or on-premises deployments keep it in your network. The terms let Permit delete data on termination, and audit log retention is by request (15 of 30). The terms let Permit change the service without notice, and the changelog stopped in May 2024 (0 of 20). Data location depends on the deployment model and is documented, but we found no subprocessor 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: permit.io, no registry record we could read (0 of 15)\n- Endpoint on the vendor's domain: {subdomain}.agent.security is not on permit.io (0 of 15)\n- Changelog: not found (0 of 10)\n- security.txt: could not be fetched (0 of 10)\n\n## 6. Security \u0026 auth, 80 out of 100, up to 3.5 more on the total\n\nWhy it scored 80: OAuth 2.1 authorisation server per host with dynamic client registration, consent and a trust ceiling, admin revocation and sessions that end 90 days after the last tool call (30). Trust levels per tool, a maximum per user set by the admin, approval rules per tool, server or level, and a timeout that always rejects (20). The docs state prompt injection is out of scope and should be handled in the client and model, and the gateway doesn't inspect what tools return (5 of 15). Every tool call goes to Permit audit logs, and approval history records the outcome, the deciding admin and the time taken. Audit log retention isn't published (13 of 15). SOC 2 Type II and HIPAA per the deployment docs. We found no disclosure policy and couldn't check security.txt (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## 7. Agent ergonomics, 83 out of 100, up to 2.8 more on the total\n\nWhy it scored 83: It adds nothing to the agent's context, since the client keeps the upstream tool list (25). The approval queue has batch actions for the admin, but nothing pages or filters for the agent (10 of 20). The agent gets a waiting message with the timeout and a rejection error carrying the admin's reason, which it can act on (18 of 20). Each tool gets a low, medium or high trust level at import, close to read-only and destructive hints, and a dropped connection cancels the request instead of leaving it to run later (15 of 20). One URL change and no SDK, with Permit SDKs in six languages for the policy side (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## 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- Where gateway incidents are reported, since the status page has no gateway component.\n- Whether the free hosted evaluation has limits or needs a card.\n- Audit log retention for gateway tool calls, which the docs say to ask Permit about.\n- Whether Enterprise agreements add notice periods for gateway changes, since the standard terms allow changes without notice.\n\n## Weaknesses\n\n- Approvals are Enterprise only, through a demo, with no published price\n- Only gateway admins approve, so routing to the right person needs admin seats\n- 5-minute default window, extendable 5 minutes at a time, suits live sessions more than overnight review\n- No gateway changelog, and the product changelog on Canny stopped in May 2024\n- Rate limits exist but aren't published, and the status page doesn't list the gateway\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- Expect a waiting message before an approval-gated tool returns, and don't retry the call while it waits\n- Read the rejection reason in the error and change approach instead of calling the same tool again\n- Treat a timeout as a rejection and ask the user to have an admin online before a batch of destructive calls\n- Stay connected while waiting, since dropping the connection cancels the request\n- On a 429 with `rate_limited`, back off for a few seconds and grow the wait on each retry\n\n## What the review panel asked for\n\n- a dated gateway changelog\n- a gateway status component\n- published audit retention\n- logged bypass use\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": 54.5,
    "assessed": "2026-10-01",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 10,
        "maxGain": 11.3,
        "reason": "No machine payment protocol (0). Approvals are Enterprise only through a demo, and the gateway has no public price (0). Evaluation starts free on the hosted gateway, but approvals aren't in it, so 10 of 20. A person signs in through the browser consent flow (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": "reliability",
        "name": "Reliability",
        "score": 47,
        "maxGain": 10.6,
        "reason": "Permit's status page at permit-io.instatus.com lists the backend, OPAL, frontend, website and PDP services, at 100 per cent, but not the gateway or agent.security (10 of 20). With no gateway component there's no incident history for it (5). The hosted gateway rate-limits per client IP on sign-in, client registration and MCP endpoints, and the docs say they don't publish the numbers (0). A 429 carries a `rate_limited` JSON body and the docs say to back off and retry with growing waits (12 of 15). Enterprise has a 99.99 per cent uptime option (10). Approvals carry an Enterprise label, not beta (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": 53,
        "maxGain": 7.6,
        "reason": "The gateway adds no tools of its own and passes upstream tool schemas through. Permit's management API has an OpenAPI-backed reference (15 of 25). No llms.txt or Markdown copies of the docs found (0). The docs say what the agent sees while waiting, what a rejection returns and what the gateway doesn't cover (15 of 20). Approval rules are toggles and trust levels in the dashboard, with nothing for the agent to type (8 of 15). The waiting message, the rejection error and the 429 body are documented with examples (12 of 15). The product changelog on Canny has no entry after May 2024, and the gateway has no changelog (3 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": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 43,
        "maxGain": 5,
        "reason": "No release notes for the gateway. The docs record new gateway capability on 2026-07-28 and 2026-07-30 (the HTTP egress proxy) and a rewrite on 17 and 20 September (20 of 30). Those are docs commits, not releases or changelog entries, so part credit for visible activity (5 of 20). Closed service with a Slack community and support email, and the public changelog stopped in May 2024 (5 of 15). Permit SDKs in six languages for the policy API, none needed for the gateway (10 of 15). Nothing to install, so we score package health on the public docs repository's CI alone (3 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"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 45,
        "maxGain": 4.8,
        "reason": "Closed service under terms updated 2026-07-01 that name Permit Inc., a Delaware corporation, under Delaware law (15 of 30). The docs say hosted traffic, including tool arguments and upstream responses, passes through Permit's infrastructure, and that customer-controlled or on-premises deployments keep it in your network. The terms let Permit delete data on termination, and audit log retention is by request (15 of 30). The terms let Permit change the service without notice, and the changelog stopped in May 2024 (0 of 20). Data location depends on the deployment model and is documented, but we found no subprocessor list (10 of 20).",
        "blend": "editorial 40, provenance 50",
        "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": "security",
        "name": "Security \u0026 auth",
        "score": 80,
        "maxGain": 3.5,
        "reason": "OAuth 2.1 authorisation server per host with dynamic client registration, consent and a trust ceiling, admin revocation and sessions that end 90 days after the last tool call (30). Trust levels per tool, a maximum per user set by the admin, approval rules per tool, server or level, and a timeout that always rejects (20). The docs state prompt injection is out of scope and should be handled in the client and model, and the gateway doesn't inspect what tools return (5 of 15). Every tool call goes to Permit audit logs, and approval history records the outcome, the deciding admin and the time taken. Audit log retention isn't published (13 of 15). SOC 2 Type II and HIPAA per the deployment docs. We found no disclosure policy and couldn't check security.txt (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": 83,
        "maxGain": 2.8,
        "reason": "It adds nothing to the agent's context, since the client keeps the upstream tool list (25). The approval queue has batch actions for the admin, but nothing pages or filters for the agent (10 of 20). The agent gets a waiting message with the timeout and a rejection error carrying the admin's reason, which it can act on (18 of 20). Each tool gets a low, medium or high trust level at import, close to read-only and destructive hints, and a dropped connection cancels the request instead of leaving it to run later (15 of 20). One URL change and no SDK, with Permit SDKs in six languages for the policy side (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"
      }
    ],
    "provenance": [
      {
        "label": "Domain age",
        "value": "permit.io, no registry record we could read",
        "points": 0,
        "max": 15
      },
      {
        "label": "Endpoint on the vendor's domain",
        "value": "{subdomain}.agent.security is not on permit.io",
        "points": 0,
        "max": 15
      },
      {
        "label": "Changelog",
        "value": "not found",
        "points": 0,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "could not be fetched",
        "points": 0,
        "max": 10
      }
    ],
    "unchecked": [
      "Where gateway incidents are reported, since the status page has no gateway component.",
      "Whether the free hosted evaluation has limits or needs a card.",
      "Audit log retention for gateway tool calls, which the docs say to ask Permit about.",
      "Whether Enterprise agreements add notice periods for gateway changes, since the standard terms allow changes without notice."
    ],
    "weaknesses": [
      "Approvals are Enterprise only, through a demo, with no published price",
      "Only gateway admins approve, so routing to the right person needs admin seats",
      "5-minute default window, extendable 5 minutes at a time, suits live sessions more than overnight review",
      "No gateway changelog, and the product changelog on Canny stopped in May 2024",
      "Rate limits exist but aren't published, and the status page doesn't list the gateway"
    ],
    "agentNotes": [
      "Expect a waiting message before an approval-gated tool returns, and don't retry the call while it waits",
      "Read the rejection reason in the error and change approach instead of calling the same tool again",
      "Treat a timeout as a rejection and ask the user to have an admin online before a batch of destructive calls",
      "Stay connected while waiting, since dropping the connection cancels the request",
      "On a 429 with `rate_limited`, back off for a few seconds and grow the wait on each retry"
    ],
    "requests": [
      {
        "text": "a dated gateway changelog",
        "reviews": 1
      },
      {
        "text": "a gateway status component",
        "reviews": 1
      },
      {
        "text": "published audit retention",
        "reviews": 1
      },
      {
        "text": "logged bypass use",
        "reviews": 1
      }
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-04",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.3",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  }
}