{
  "fixes": {
    "slug": "infisical",
    "name": "Infisical",
    "listing": "https://www.anchorterminal.com/tools/infisical",
    "markdown": "# Fix list: Infisical\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/infisical, the October 2026 research run, assessed 1 October 2026. Grade A, 81.9 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 Infisical: 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: No x402, MPP or L402 (0). Pro $20 and Advanced $40 per identity a month billed yearly are public, Enterprise is custom, and there's no per-call price (10). Free plan with 5 identities, and the pricing page says no credit card for Free or for the Pro and Advanced trials (20). A person signs up in a browser before any machine identity exists, and nothing lets an agent create its own account (0). The MIT core self-hosts free, but under the rubric we score the hosted option.\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. Schema \u0026 documentation, 87 out of 100, up to 2.1 more on the total\n\nWhy it scored 87: Every instance serves an OpenAPI document at /api/docs/json, generated from the Zod route schemas and published by a CI workflow (25). llms.txt at infisical.com/docs per the 30 September check, Mintlify docs (10). The API reference describes each parameter, but the official MCP server's tool descriptions are one line each (\"Create a new secret in Infisical\") with nothing on when not to use them (12 of 20). Typed parameters with required fields and defaults in the OpenAPI and in the 10 MCP input schemas (13 of 15). A documented error format with a stable error identifier, a request ID and a list of status codes, and examples on the reference pages (15). Endpoints are versioned per path (v1, v3, v4) and every release gets an upgrade-impact file listing breaking changes, but the docs changelog stops at July 2025 (12 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## 3. Reliability, 90 out of 100, up to 2 more on the total\n\nWhy it scored 90: Status page at status.infisical.com (incident.io) with monthly history (20). Since 3 July 2026 the only entry is a planned Entitlement Service upgrade on 23 July, and August and September read no incidents reported (30). Cloud rate limits published per plan and per client IP, 600 a minute overall and 200 reads, 90 writes and 120 secret operations a minute on Free (15). The 429 body says how many seconds remain, and the errors page says to retry GET, PUT and DELETE with exponential backoff on a 5xx and not to blindly retry a POST or PATCH (15). No SLA found on the pricing page or in the docs (0). The secrets API and machine identities are generally available (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## 4. Security \u0026 auth, 91 out of 100, up to 1.6 more on the total\n\nWhy it scored 91: Machine identities log in with one of 13 methods (Universal Auth, OIDC, JWT, AWS, Azure, GCP, Kubernetes, SPIFFE and others) and get a short-lived access token with a TTL, revocable at /api/v1/auth/token/revoke, with project roles scoped by environment and path (30). Custom roles down to read-only on one path, change requests and access requests with approvals, and an MCP allowlist that can expose only list and get tools (20). Agent Vault and Agent Proxy attach the credential at the proxy so the model never holds it, and INFISICAL_MASK_SECRET_VALUES replaces values with \u003cmasked\u003e in MCP responses, though masking is off by default (13 of 15). Audit logs on Pro (30 days) and Advanced (90 days), none on Free, and Agent Vault session logs go to an S3 bucket you own (12 of 15). security.txt valid to 2027-08-01, a Bugcrowd disclosure programme with a 3-business-day acknowledgement, a private paid bounty, and SECURITY.md routes SOC 2 report requests to security@infisical.com. No advisories are published on the GitHub repository, so we couldn't judge how past ones were handled (16 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## 5. Agent ergonomics, 91 out of 100, up to 1.5 more on the total\n\nWhy it scored 91: The v4 secrets list takes viewSecretValue=false to return names without values, plus tagSlugs, metadataFilter and recursive, and the OpenAPI can be trimmed with ?tag=secrets. The official MCP server has 10 tools and an INFISICAL_ENABLED_TOOLS allowlist (23 of 25). List endpoints page with offset, limit and totalCount, and secrets filter by path, tag and metadata, though the secrets list itself isn't paged (18 of 20). Errors carry a machine-readable error class (NotFound, PermissionDenied, RateLimitExceeded) and a reqId (20). The MCP tools carry readOnlyHint, destructiveHint and idempotentHint, and the docs give safe-retry rules, but there are no idempotency keys for POST (16 of 20). SDKs for Node, Python, Go, Java, .NET, Ruby, PHP, C++ and Rust; secretPath defaults to /, while projectId and environment are always required (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## 6. Transparency \u0026 trust, 85 out of 100, up to 1.3 more on the total\n\nMade of editorial 77, provenance 92.\n\nWhy it scored 85: MIT outside the ee/ directories, with Agent Vault, audit log streaming and other paid modules under a proprietary ee licence (24 of 30). Privacy policy (15 September 2025) and a subprocessor list (9 September 2026, 17 entries) agree, and a DPA sits in the terms hub per the 30 September check, but retention is only as long as necessary and every subprocessor is listed in the United States although an EU region is sold (20 of 30). Native Integrations retire on 19 August 2027 with a migration guide, and each release gets an upgrade-impact file, but there's no general deprecation policy (15 of 20). Self-hosted telemetry is on by default with TELEMETRY_ENABLED=false documented as the opt-out, and PostHog is on the subprocessor list (18 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: infisical.com, registered 2022-07-06 (4 years) (7 of 15)\n\n## 7. Maintenance \u0026 community, 90 out of 100, up to 0.9 more on the total\n\nWhy it scored 90: v0.165.16 tagged on 23 September 2026, 8 days before this check (30). 48 tagged releases between 3 July and 23 September 2026, and the MCP server shipped 0.0.24 on 9 September (20). 262 open issues and commits merged daily, but the issue we sampled (#8052) had a reply from a third-party support bot rather than a maintainer (15 of 25). Current official SDKs in nine languages (15). GitHub Actions pinned by commit SHA, dependency audit enforced in the MCP server, backend, Go and Helm test workflows (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## 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- The listing said there was no official MCP server for secret operations; @infisical/mcp exists (10 tools, 0.0.24 on 9 September 2026), so the notable, details, connect, packages and toolCount fields are corrected in the patch.\n- Whether a paid plan carries an SLA; none was found on the pricing page or in the docs.\n- Whether the 429 also sends a Retry-After header; the rate limiter's code builds the message but we didn't run it.\n- unchecked: llms.txt, relied on from the 30 September check.\n- How quickly maintainers answer issues; the one we sampled had only a third-party bot reply.\n\n## Weaknesses\n\n- Free has no audit logs, Pro keeps them 30 days, and dynamic secrets need Advanced at $40 an identity a month\n- Cloud rate limits are per client IP, so agents behind one NAT share 600 requests a minute\n- The MCP server returns secret values unless INFISICAL_MASK_SECRET_VALUES is set, and it's still version 0.0.x\n- Agent Vault session tokens travel to the proxy unencrypted, and the feature sits under the proprietary ee/ licence\n- No SLA found, and every listed subprocessor is in the United States despite the EU region\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- Run a coding agent under `infisical agent-vault run` with a bundle that allows only the hosts, methods and paths it needs, and set --ttl to the job length\n- Start @infisical/mcp with INFISICAL_ENABLED_TOOLS=list-projects,list-secrets,get-secret and INFISICAL_MASK_SECRET_VALUES=true unless the model must see a value\n- Log in once with Universal Auth and keep the access token for its TTL, since identity logins count against the per-IP write limit\n- Pass viewSecretValue=false to GET /api/v4/secrets when you only need names, and expandSecretReferences=true when values reference other secrets\n- On a 429 read the seconds from the message field and wait that long; don't retry a POST after a 5xx without checking it didn't land\n\n## What the review panel asked for\n\n- Agent self-signup\n- Agent Vault plan gating\n- Masking on by default\n- Per-identity rate limits\n- Audit logs on Free\n- Say when not to use each tool in its description\n- Turn value masking on by default\n- resume the docs changelog\n- Idempotency keys on POST\n- Confirm whether `Retry-After` is sent\n- semver matching impact files\n- mask values by default\n- publish past advisories\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": "A",
    "score": 81.9,
    "assessed": "2026-10-01",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 30,
        "maxGain": 8.8,
        "reason": "No x402, MPP or L402 (0). Pro $20 and Advanced $40 per identity a month billed yearly are public, Enterprise is custom, and there's no per-call price (10). Free plan with 5 identities, and the pricing page says no credit card for Free or for the Pro and Advanced trials (20). A person signs up in a browser before any machine identity exists, and nothing lets an agent create its own account (0). The MIT core self-hosts free, but under the rubric we score the hosted option.",
        "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": "schema",
        "name": "Schema \u0026 documentation",
        "score": 87,
        "maxGain": 2.1,
        "reason": "Every instance serves an OpenAPI document at /api/docs/json, generated from the Zod route schemas and published by a CI workflow (25). llms.txt at infisical.com/docs per the 30 September check, Mintlify docs (10). The API reference describes each parameter, but the official MCP server's tool descriptions are one line each (\"Create a new secret in Infisical\") with nothing on when not to use them (12 of 20). Typed parameters with required fields and defaults in the OpenAPI and in the 10 MCP input schemas (13 of 15). A documented error format with a stable error identifier, a request ID and a list of status codes, and examples on the reference pages (15). Endpoints are versioned per path (v1, v3, v4) and every release gets an upgrade-impact file listing breaking changes, but the docs changelog stops at July 2025 (12 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": "reliability",
        "name": "Reliability",
        "score": 90,
        "maxGain": 2,
        "reason": "Status page at status.infisical.com (incident.io) with monthly history (20). Since 3 July 2026 the only entry is a planned Entitlement Service upgrade on 23 July, and August and September read no incidents reported (30). Cloud rate limits published per plan and per client IP, 600 a minute overall and 200 reads, 90 writes and 120 secret operations a minute on Free (15). The 429 body says how many seconds remain, and the errors page says to retry GET, PUT and DELETE with exponential backoff on a 5xx and not to blindly retry a POST or PATCH (15). No SLA found on the pricing page or in the docs (0). The secrets API and machine identities are generally available (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": "security",
        "name": "Security \u0026 auth",
        "score": 91,
        "maxGain": 1.6,
        "reason": "Machine identities log in with one of 13 methods (Universal Auth, OIDC, JWT, AWS, Azure, GCP, Kubernetes, SPIFFE and others) and get a short-lived access token with a TTL, revocable at /api/v1/auth/token/revoke, with project roles scoped by environment and path (30). Custom roles down to read-only on one path, change requests and access requests with approvals, and an MCP allowlist that can expose only list and get tools (20). Agent Vault and Agent Proxy attach the credential at the proxy so the model never holds it, and INFISICAL_MASK_SECRET_VALUES replaces values with \u003cmasked\u003e in MCP responses, though masking is off by default (13 of 15). Audit logs on Pro (30 days) and Advanced (90 days), none on Free, and Agent Vault session logs go to an S3 bucket you own (12 of 15). security.txt valid to 2027-08-01, a Bugcrowd disclosure programme with a 3-business-day acknowledgement, a private paid bounty, and SECURITY.md routes SOC 2 report requests to security@infisical.com. No advisories are published on the GitHub repository, so we couldn't judge how past ones were handled (16 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": 91,
        "maxGain": 1.5,
        "reason": "The v4 secrets list takes viewSecretValue=false to return names without values, plus tagSlugs, metadataFilter and recursive, and the OpenAPI can be trimmed with ?tag=secrets. The official MCP server has 10 tools and an INFISICAL_ENABLED_TOOLS allowlist (23 of 25). List endpoints page with offset, limit and totalCount, and secrets filter by path, tag and metadata, though the secrets list itself isn't paged (18 of 20). Errors carry a machine-readable error class (NotFound, PermissionDenied, RateLimitExceeded) and a reqId (20). The MCP tools carry readOnlyHint, destructiveHint and idempotentHint, and the docs give safe-retry rules, but there are no idempotency keys for POST (16 of 20). SDKs for Node, Python, Go, Java, .NET, Ruby, PHP, C++ and Rust; secretPath defaults to /, while projectId and environment are always required (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": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 85,
        "maxGain": 1.3,
        "reason": "MIT outside the ee/ directories, with Agent Vault, audit log streaming and other paid modules under a proprietary ee licence (24 of 30). Privacy policy (15 September 2025) and a subprocessor list (9 September 2026, 17 entries) agree, and a DPA sits in the terms hub per the 30 September check, but retention is only as long as necessary and every subprocessor is listed in the United States although an EU region is sold (20 of 30). Native Integrations retire on 19 August 2027 with a migration guide, and each release gets an upgrade-impact file, but there's no general deprecation policy (15 of 20). Self-hosted telemetry is on by default with TELEMETRY_ENABLED=false documented as the opt-out, and PostHog is on the subprocessor list (18 of 20).",
        "blend": "editorial 77, provenance 92",
        "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": 90,
        "maxGain": 0.9,
        "reason": "v0.165.16 tagged on 23 September 2026, 8 days before this check (30). 48 tagged releases between 3 July and 23 September 2026, and the MCP server shipped 0.0.24 on 9 September (20). 262 open issues and commits merged daily, but the issue we sampled (#8052) had a reply from a third-party support bot rather than a maintainer (15 of 25). Current official SDKs in nine languages (15). GitHub Actions pinned by commit SHA, dependency audit enforced in the MCP server, backend, Go and Helm test workflows (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": "infisical.com, registered 2022-07-06 (4 years)",
        "points": 7,
        "max": 15
      }
    ],
    "unchecked": [
      "The listing said there was no official MCP server for secret operations; @infisical/mcp exists (10 tools, 0.0.24 on 9 September 2026), so the notable, details, connect, packages and toolCount fields are corrected in the patch.",
      "Whether a paid plan carries an SLA; none was found on the pricing page or in the docs.",
      "Whether the 429 also sends a Retry-After header; the rate limiter's code builds the message but we didn't run it.",
      "unchecked: llms.txt, relied on from the 30 September check.",
      "How quickly maintainers answer issues; the one we sampled had only a third-party bot reply."
    ],
    "weaknesses": [
      "Free has no audit logs, Pro keeps them 30 days, and dynamic secrets need Advanced at $40 an identity a month",
      "Cloud rate limits are per client IP, so agents behind one NAT share 600 requests a minute",
      "The MCP server returns secret values unless INFISICAL_MASK_SECRET_VALUES is set, and it's still version 0.0.x",
      "Agent Vault session tokens travel to the proxy unencrypted, and the feature sits under the proprietary ee/ licence",
      "No SLA found, and every listed subprocessor is in the United States despite the EU region"
    ],
    "agentNotes": [
      "Run a coding agent under `infisical agent-vault run` with a bundle that allows only the hosts, methods and paths it needs, and set --ttl to the job length",
      "Start @infisical/mcp with INFISICAL_ENABLED_TOOLS=list-projects,list-secrets,get-secret and INFISICAL_MASK_SECRET_VALUES=true unless the model must see a value",
      "Log in once with Universal Auth and keep the access token for its TTL, since identity logins count against the per-IP write limit",
      "Pass viewSecretValue=false to GET /api/v4/secrets when you only need names, and expandSecretReferences=true when values reference other secrets",
      "On a 429 read the seconds from the message field and wait that long; don't retry a POST after a 5xx without checking it didn't land"
    ],
    "requests": [
      {
        "text": "Agent self-signup",
        "reviews": 1
      },
      {
        "text": "Agent Vault plan gating",
        "reviews": 1
      },
      {
        "text": "Masking on by default",
        "reviews": 1
      },
      {
        "text": "Per-identity rate limits",
        "reviews": 1
      },
      {
        "text": "Audit logs on Free",
        "reviews": 1
      },
      {
        "text": "Say when not to use each tool in its description",
        "reviews": 1
      },
      {
        "text": "Turn value masking on by default",
        "reviews": 1
      },
      {
        "text": "resume the docs changelog",
        "reviews": 1
      },
      {
        "text": "Idempotency keys on POST",
        "reviews": 1
      },
      {
        "text": "Confirm whether `Retry-After` is sent",
        "reviews": 1
      },
      {
        "text": "semver matching impact files",
        "reviews": 1
      },
      {
        "text": "mask values by default",
        "reviews": 1
      },
      {
        "text": "publish past advisories",
        "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"
  }
}