{
  "fixes": {
    "slug": "launchdarkly",
    "name": "LaunchDarkly",
    "listing": "https://www.anchorterminal.com/tools/launchdarkly",
    "markdown": "# Fix list: LaunchDarkly\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/launchdarkly, the October 2026 research run, assessed 8 October 2026. Grade B, 69.2 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 LaunchDarkly: 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, 40 out of 100, up to 7.5 more on the total\n\nWhy it scored 40: No x402, MPP or L402 (0). Per-unit prices are public without a login, at $10 a service connection a month, $8.33 per 1,000 client-side MAU a month billed yearly and $5 per 1,000 additional AI runs. Enterprise is custom (20). The Developer plan is free with no card (20). A person signs up in a browser and approves OAuth or creates a token (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, 64 out of 100, up to 7.2 more on the total\n\nWhy it scored 64: Graded on the hosted service with the hosted lines. Statuspage at status.launchdarkly.com with 51 components and incident history (20). 13 incidents between 10 July and 8 October 2026. On 10 July the web application was unavailable and flag delivery failed, in two incidents from 09:16 to 11:19 PT marked critical, with a follow-up on SDK 401s open until 29 July. On 2 October elevated errors on the application and API stayed open for 18 hours 55 minutes, marked minor. One major outage, so 10 of 30. The API docs describe global, route, token and IP limits and say the numbers are not published. Limit and remaining headers report them at run time (5 of 15). 429s carry reset headers, IP limits carry `Retry-After`, and the docs ask for jitter and backoff. No idempotency keys were found, though JSON patch can test a resource version and 409 is documented as retryable (11 of 15). The SLA commits to 99.9 per cent for Enterprise support and 99.99 per cent for Premium support (10). The REST API and hosted MCP server carry no beta label, but 132 of 401 API operations are beta (8 of 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. Security \u0026 auth, 66 out of 100, up to 6 more on the total\n\nWhy it scored 66: The hosted MCP server uses OAuth with PKCE (S256), dynamic client registration, a revocation endpoint and three scopes, and an authorised app can never do more than the member who approved it. API access tokens are revocable and resettable with a Reader, Writer, Admin or Owner role. Custom roles and inline policies need a plan with custom roles, and service tokens need Enterprise. No secret in a URL query string is documented (26 of 30). Reader role and reader scope, `--scope read` and `--tool` on the local server, and approval requests for changes on Enterprise. No confirmation step for destructive hosted tools was found (14 of 20). The observability tools return logs, traces, errors and session data written by end users. No prompt-injection guidance was found in the MCP docs, which warn only that an API key in a config file may appear in a conversation (3 of 15). Change history with an API at `/api/v2/auditlog` and comments on updates, limited to 30 days on the Developer plan, and the pricing table lists audit logging under Enterprise only (11 of 15). The security page lists SOC 2 Type II, ISO 27001 and FedRAMP authorisation. security.txt returns 404, and no disclosure policy or bug bounty was found on the pages read (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## 4. Agent ergonomics, 69 out of 100, up to 5 more on the total\n\nWhy it scored 69: The hosted MCP server lists 141 tools, far past 30 (5). The OAuth metadata names reader, writer and observability scopes, but no tool filter for the hosted server was found. The `--tool` and `--scope read` options belong to the local server, which has 20 tools, so 3 of 10 added back (8). The API has `limit` and `offset` on 61 operations, filters and sorting, summary and detailed representations and an `expand` parameter on 40 (18). Errors carry `code`, `message` and `id`, with a table of statuses and fixes (15). The server page marks 73 tools read-only, 35 destructive and 89 idempotent. There are no idempotency keys on REST writes, and JSON patch can test a version before applying (15). Generated API clients on npm (`launchdarkly-api` 20.0.0) and PyPI (24.0.0) and the `ldcli` CLI, with semantic patch needing a special content type (13).\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## 5. Schema \u0026 documentation, 89 out of 100, up to 1.8 more on the total\n\nWhy it scored 89: OpenAPI 3.0.3 served without a login at app.launchdarkly.com/api/v2/openapi.json, with 253 paths and 401 operations (25). The hosted MCP input schemas need a sign-in and were not read. A docs index for agents at launchdarkly.com/docs/llms.txt, and docs pages answer as Markdown with .md (10). 399 of 401 operations have a description. The MCP tool descriptions on the server page average about 740 characters and state pagination, sorting, plan limits and which tool to call next (17). 169 enums across 746 schemas. JSON patch bodies and semantic patch instructions are loosely typed objects (9). 1,853 examples in the spec, 429 documented on 287 operations and 403 on 355, and an error table with a suggested fix for each status (13). Dated API versions chosen by header or per token, a version changelog with end-of-life dates and a dated docs changelog (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, 80 out of 100, up to 1.8 more on the total\n\nMade of editorial 74, provenance 86.\n\nWhy it scored 80: Closed hosted service under published terms. The SDKs and `ldcli` are Apache-2.0 and the local MCP server is MIT (17 of 30). The privacy policy (8 April 2025) excludes data processed for customers, which the DPA (1 May 2024) covers, with deletion on written request after termination. The privacy policy gives no retention periods. The terms allow use of anonymised and aggregated data, and no clause on model training was found (20 of 30). The end-of-life policy gives API and SDK versions 12 months after a new release, and the version table lists end-of-life dates. Beta resources may change without notice (17 of 20). The sub-processor list (15 May 2026) gives locations, with Anthropic, OpenAI and Speakeasy under AI services, the DPA gives 30 days' notice of new sub-processors, and EU and federal instances exist (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- Terms of service: read, states 6 of the 7 things a reader expects, and has 1 clause that costs points (7.1 of 10)\n- Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10)\n- security.txt: not found (0 of 10)\n\n## 7. Maintenance \u0026 community, 82 out of 100, up to 1.6 more on the total\n\nWhy it scored 82: The Python server SDK 9.18.1 was published on 6 October 2026, the local MCP server v0.7.0 on 5 October and the docs changelog's newest entries are 7 October (30). The docs changelog has 50 dated entries since 3 August 2026, and `ldcli` had eight releases between 28 July and 24 September (20). Public docs changelog, a support site and a Discord community named on the pricing page. We did not read how quickly GitHub issues are answered (10 of 25). Current official SDKs and generated API clients. The official MCP registry has no LaunchDarkly namespace, only two third-party entries (15). The MCP server repository has four CI workflows and Dependabot fixes merged on 18 August 2026, with three commits since 10 July, and its README says the local server is updated less often than the hosted one (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## What we couldn't check\n\nWhat we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.\n\n- unchecked: the live `tools/list` response of the hosted MCP server, which needs an OAuth sign-in. Tool names, descriptions and annotations come from the server page, which states 141 tools, and input schemas were not read\n- unchecked: the trust centre, which redirects to app.vanta.com and was not read, so a disclosure policy or bug bounty may exist there\n- unchecked: GitHub star counts and issue response times, because the GitHub API answered with a rate limit\n- unchecked: what the reader, writer and observability OAuth scopes each allow. They appear in the server metadata and are not described in the docs we read\n- No price for experimentation above the 100,000 MAU included was found on the pricing page\n- The pricing table lists audit logging under Enterprise only, while the docs say change history is on every plan with 30 days on Developer. We scored the docs' statement\n- The hosted MCP endpoint is on mcp.launchdarkly.com and its response headers name Gram, a Speakeasy product. Speakeasy is on the sub-processor list\n- The privacy policy says it does not apply to data LaunchDarkly processes for customers. The DPA covers that data and is linked in the notes\n- The category was kept as product-analytics. LaunchDarkly covers its flags and experiments half and has no funnel, retention or cohort queries\n\n## Weaknesses\n\n- The hosted MCP server lists 141 tools, and no tool filter for it was found in the reviewed documentation\n- The API docs say the numeric rate limits are not published and may change\n- 132 of 401 API operations are beta, need `LD-API-Version: beta` and may change without notice\n- On 10 July 2026 the web application was unavailable and flag delivery failed for part of two hours, and some SDKs needed a restart\n- Custom roles, inline token policies, service tokens and approvals need an Enterprise plan, and the hosted MCP server is absent from the EU and federal instances\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- Connect to https://mcp.launchdarkly.com/mcp/launchdarkly over streamable HTTP with OAuth. On the EU or federal instance run the local `@launchdarkly/mcp-server` with `--server-url`\n- Authorise with a Reader role or the `reader` scope unless the task writes. A 403 after consent means the member's role lacks access\n- Send `LD-API-Version: 20240415` on every REST call, and `LD-API-Version: beta` for beta resources, which answer 403 without it\n- Read `X-Ratelimit-Global-Remaining`, `X-Ratelimit-Route-Remaining` and `X-Ratelimit-Reset` on each response, and wait `Retry-After` seconds on an IP limit\n- Send `Content-Type: application/json; domain-model=launchdarkly.semanticpatch` for semantic patches, or the body is read as JSON patch and answers 400\n\n## When it's done\n\nSend what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `\"kind\": \"dispute\"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.\n",
    "grade": "B",
    "score": 69.2,
    "assessed": "2026-10-08",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 40,
        "maxGain": 7.5,
        "reason": "No x402, MPP or L402 (0). Per-unit prices are public without a login, at $10 a service connection a month, $8.33 per 1,000 client-side MAU a month billed yearly and $5 per 1,000 additional AI runs. Enterprise is custom (20). The Developer plan is free with no card (20). A person signs up in a browser and approves OAuth or creates a token (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": 64,
        "maxGain": 7.2,
        "reason": "Graded on the hosted service with the hosted lines. Statuspage at status.launchdarkly.com with 51 components and incident history (20). 13 incidents between 10 July and 8 October 2026. On 10 July the web application was unavailable and flag delivery failed, in two incidents from 09:16 to 11:19 PT marked critical, with a follow-up on SDK 401s open until 29 July. On 2 October elevated errors on the application and API stayed open for 18 hours 55 minutes, marked minor. One major outage, so 10 of 30. The API docs describe global, route, token and IP limits and say the numbers are not published. Limit and remaining headers report them at run time (5 of 15). 429s carry reset headers, IP limits carry `Retry-After`, and the docs ask for jitter and backoff. No idempotency keys were found, though JSON patch can test a resource version and 409 is documented as retryable (11 of 15). The SLA commits to 99.9 per cent for Enterprise support and 99.99 per cent for Premium support (10). The REST API and hosted MCP server carry no beta label, but 132 of 401 API operations are beta (8 of 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": 66,
        "maxGain": 6,
        "reason": "The hosted MCP server uses OAuth with PKCE (S256), dynamic client registration, a revocation endpoint and three scopes, and an authorised app can never do more than the member who approved it. API access tokens are revocable and resettable with a Reader, Writer, Admin or Owner role. Custom roles and inline policies need a plan with custom roles, and service tokens need Enterprise. No secret in a URL query string is documented (26 of 30). Reader role and reader scope, `--scope read` and `--tool` on the local server, and approval requests for changes on Enterprise. No confirmation step for destructive hosted tools was found (14 of 20). The observability tools return logs, traces, errors and session data written by end users. No prompt-injection guidance was found in the MCP docs, which warn only that an API key in a config file may appear in a conversation (3 of 15). Change history with an API at `/api/v2/auditlog` and comments on updates, limited to 30 days on the Developer plan, and the pricing table lists audit logging under Enterprise only (11 of 15). The security page lists SOC 2 Type II, ISO 27001 and FedRAMP authorisation. security.txt returns 404, and no disclosure policy or bug bounty was found on the pages read (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": 69,
        "maxGain": 5,
        "reason": "The hosted MCP server lists 141 tools, far past 30 (5). The OAuth metadata names reader, writer and observability scopes, but no tool filter for the hosted server was found. The `--tool` and `--scope read` options belong to the local server, which has 20 tools, so 3 of 10 added back (8). The API has `limit` and `offset` on 61 operations, filters and sorting, summary and detailed representations and an `expand` parameter on 40 (18). Errors carry `code`, `message` and `id`, with a table of statuses and fixes (15). The server page marks 73 tools read-only, 35 destructive and 89 idempotent. There are no idempotency keys on REST writes, and JSON patch can test a version before applying (15). Generated API clients on npm (`launchdarkly-api` 20.0.0) and PyPI (24.0.0) and the `ldcli` CLI, with semantic patch needing a special content type (13).",
        "checklist": [
          "- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.",
          "Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-ergonomics"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 89,
        "maxGain": 1.8,
        "reason": "OpenAPI 3.0.3 served without a login at app.launchdarkly.com/api/v2/openapi.json, with 253 paths and 401 operations (25). The hosted MCP input schemas need a sign-in and were not read. A docs index for agents at launchdarkly.com/docs/llms.txt, and docs pages answer as Markdown with .md (10). 399 of 401 operations have a description. The MCP tool descriptions on the server page average about 740 characters and state pagination, sorting, plan limits and which tool to call next (17). 169 enums across 746 schemas. JSON patch bodies and semantic patch instructions are loosely typed objects (9). 1,853 examples in the spec, 429 documented on 287 operations and 403 on 355, and an error table with a suggested fix for each status (13). Dated API versions chosen by header or per token, a version changelog with end-of-life dates and a dated docs changelog (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": 80,
        "maxGain": 1.8,
        "reason": "Closed hosted service under published terms. The SDKs and `ldcli` are Apache-2.0 and the local MCP server is MIT (17 of 30). The privacy policy (8 April 2025) excludes data processed for customers, which the DPA (1 May 2024) covers, with deletion on written request after termination. The privacy policy gives no retention periods. The terms allow use of anonymised and aggregated data, and no clause on model training was found (20 of 30). The end-of-life policy gives API and SDK versions 12 months after a new release, and the version table lists end-of-life dates. Beta resources may change without notice (17 of 20). The sub-processor list (15 May 2026) gives locations, with Anthropic, OpenAI and Speakeasy under AI services, the DPA gives 30 days' notice of new sub-processors, and EU and federal instances exist (20).",
        "blend": "editorial 74, provenance 86",
        "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": 82,
        "maxGain": 1.6,
        "reason": "The Python server SDK 9.18.1 was published on 6 October 2026, the local MCP server v0.7.0 on 5 October and the docs changelog's newest entries are 7 October (30). The docs changelog has 50 dated entries since 3 August 2026, and `ldcli` had eight releases between 28 July and 24 September (20). Public docs changelog, a support site and a Discord community named on the pricing page. We did not read how quickly GitHub issues are answered (10 of 25). Current official SDKs and generated API clients. The official MCP registry has no LaunchDarkly namespace, only two third-party entries (15). The MCP server repository has four CI workflows and Dependabot fixes merged on 18 August 2026, with three commits since 10 July, and its README says the local server is updated less often than the hosted one (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": "Terms of service",
        "value": "read, states 6 of the 7 things a reader expects, and has 1 clause that costs points",
        "points": 7.1,
        "max": 10
      },
      {
        "label": "Privacy policy",
        "value": "read, states 7 of the 8 things a reader expects",
        "points": 9.3,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "not found",
        "points": 0,
        "max": 10
      }
    ],
    "unchecked": [
      "unchecked: the live `tools/list` response of the hosted MCP server, which needs an OAuth sign-in. Tool names, descriptions and annotations come from the server page, which states 141 tools, and input schemas were not read",
      "unchecked: the trust centre, which redirects to app.vanta.com and was not read, so a disclosure policy or bug bounty may exist there",
      "unchecked: GitHub star counts and issue response times, because the GitHub API answered with a rate limit",
      "unchecked: what the reader, writer and observability OAuth scopes each allow. They appear in the server metadata and are not described in the docs we read",
      "No price for experimentation above the 100,000 MAU included was found on the pricing page",
      "The pricing table lists audit logging under Enterprise only, while the docs say change history is on every plan with 30 days on Developer. We scored the docs' statement",
      "The hosted MCP endpoint is on mcp.launchdarkly.com and its response headers name Gram, a Speakeasy product. Speakeasy is on the sub-processor list",
      "The privacy policy says it does not apply to data LaunchDarkly processes for customers. The DPA covers that data and is linked in the notes",
      "The category was kept as product-analytics. LaunchDarkly covers its flags and experiments half and has no funnel, retention or cohort queries"
    ],
    "weaknesses": [
      "The hosted MCP server lists 141 tools, and no tool filter for it was found in the reviewed documentation",
      "The API docs say the numeric rate limits are not published and may change",
      "132 of 401 API operations are beta, need `LD-API-Version: beta` and may change without notice",
      "On 10 July 2026 the web application was unavailable and flag delivery failed for part of two hours, and some SDKs needed a restart",
      "Custom roles, inline token policies, service tokens and approvals need an Enterprise plan, and the hosted MCP server is absent from the EU and federal instances"
    ],
    "agentNotes": [
      "Connect to https://mcp.launchdarkly.com/mcp/launchdarkly over streamable HTTP with OAuth. On the EU or federal instance run the local `@launchdarkly/mcp-server` with `--server-url`",
      "Authorise with a Reader role or the `reader` scope unless the task writes. A 403 after consent means the member's role lacks access",
      "Send `LD-API-Version: 20240415` on every REST call, and `LD-API-Version: beta` for beta resources, which answer 403 without it",
      "Read `X-Ratelimit-Global-Remaining`, `X-Ratelimit-Route-Remaining` and `X-Ratelimit-Reset` on each response, and wait `Retry-After` seconds on an IP limit",
      "Send `Content-Type: application/json; domain-model=launchdarkly.semanticpatch` for semantic patches, or the body is read as JSON patch and answers 400"
    ],
    "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"
  }
}