# Fix list: Pirate Weather From Anchor Terminal's listing at https://www.anchorterminal.com/tools/pirate-weather, the October 2026 research run, assessed 1 October 2026. Grade B, 66.2 out of 100. This 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. For a coding agent working on Pirate Weather: 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. ## 1. Security & auth, 45 out of 100, up to 9.6 more on the total Why it scored 45: A plain API key from the Apiable portal. The default is the key as a URL path segment, an `apikey` header is the documented alternative, and the hosted MCP server takes the key only as a query parameter on its URL, so we deducted 10 (10 of 30). Read-only data, no write actions (15 of 20). Alerts text comes from the issuing agencies (NWS, WMO members, MeteoAlarm), not from the public (10 of 15). Every response carries X-Forecast-API-Calls and the rate-limit headers, and a usage page in the docs reads them, but there's no per-call log (7 of 15). No security.txt (404 per the 30 September check) and no SECURITY.md in either repository. CodeQL and Dependabot run on the code repository. No bounty or certification (3 of 20). The checklist (https://www.anchorterminal.com/benchmark/#checklist-security): - 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. - 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions. - 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. - 0 to 15, audit logs or per-call visibility for the operator. - 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. ## 2. Payments & pricing, 30 out of 100, up to 8.8 more on the total Why it scored 30: No x402, MPP or L402 (0). The free tier and the $2 plan are published, but higher plans sit in a portal that renders only with JavaScript, so we count plan pricing that's partly public (10 of 20). 10,000 calls a month free with no card, per the 30 September check (20). A person has to sign up at pirate-weather.apiable.io and subscribe, and the key can take 20 minutes to go live (0). The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments): 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. - 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. - 20, a free tier or trial that doesn't need a card. - 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. ## 3. Reliability, 75 out of 100, up to 5 more on the total Why it scored 75: Status page at pirateweather.xitoring.io, linked from the docs. It renders only with JavaScript, so we relied on the 30 September check that it exists (20). The project's own incident reports page is readable. In the last 90 days it records one incident, on 27 August 2026, when a cleanup job at the Apiable portal removed about 140 subscriptions and revoked their keys while the API itself stayed up (20). Limits published with numbers, 10,000 calls a month free and 20,000 on the $2 plan, with Ratelimit-Limit, Ratelimit-Remaining and Ratelimit-Reset headers (15). 429 means the monthly quota is spent and the reset header says when it clears, and the docs say to retry a 500. No backoff guidance beyond that, though every call is a GET (10 of 15). The terms say best effort with no warranty, so no SLA (0). Version 2 of the API is the stable default (10). The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability): Hosted APIs, MCP servers, models and platforms. - 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own). - 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. - 15, rate limits documented with numbers. - 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved. - 10, an SLA published for any paid tier. - 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. - 25, a public CI and test suite, passing on the default branch. - 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered). - 15, semver discipline and breaking changes called out in a changelog. - 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. ## 4. Agent ergonomics, 77 out of 100, up to 3.7 more on the total Why it scored 77: `exclude`, `blocks`, `hourly_indices`, `daily_indices` and `day_night_indices` cut a response down to the rows asked for, and the MCP `get_forecast` tool returns a compact this-hour and today summary. Eleven MCP tools, so a little over the ten-tool line (22 of 25). Output-size controls cover what pagination would, since each call is one location (18 of 20). Error codes are documented with causes, and MCP tools return `ok: false` with the HTTP status and message, but there's no machine-readable error code in the body (15 of 20). Read-only GETs. The MCP tools carry no readOnlyHint annotations (14 of 20). One required argument, the coordinate or a city and country. No official SDK, though Dark Sky client libraries work against it (8 of 15). The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics): - 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). - 20, pagination, filtering and output-size controls. - 20, actionable, documented error responses, codes and messages an agent can recover from. - 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations. - 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. ## 5. Transparency & trust, 71 out of 100, up to 2.5 more on the total Made of editorial 55, provenance 86. Why it scored 71: The API code is AGPL-3.0 and the docs and OpenAPI spec are Apache-2.0. We took 5 off the 30 because the terms say nothing about reusing response data (25 of 30). The privacy notice is a generator template last updated 23 September 2024, with retention 'as long as necessary' and no DPA. It agrees with itself but says little (12 of 30). Version 1 and 2 response shapes are both kept, and the changelog dates every change, but there's no written deprecation policy (10 of 20). The privacy notice names Apiable, and the infrastructure blog posts name AWS ECS, Aurora PostgreSQL and Kong, but no data locations are stated (8 of 20). The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency): - 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. - 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors). - 0 to 20, a deprecation policy or notices with dates. - 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. Provenance checks not met in full (half of this category, computed from checked facts): - Domain age: pirateweather.net, registered 2021-03-01 (5 years) (11 of 15) - security.txt: not found (0 of 10) ## 6. Schema & documentation, 85 out of 100, up to 2.4 more on the total Why it scored 85: OpenAPI 3.1.0 spec in the docs repository, versioned 2.10.2 to match the API, with 18 parameters and 400, 401, 404, 429, 500 and 502 responses on the forecast path (25). No llms.txt (404), though the MkDocs Markdown sources are public on GitHub (5 of 10). The API docs and data-sources page explain each block and which model fills each field. The 11 MCP tool docstrings are one line each, with no when-not-to-use (14 of 20). The spec has five enums and leaves `exclude` and `include` as comma strings. The MCP tools use Pydantic types with bounded latitude, longitude and day counts and enumerated units and languages (11 of 15). A full response example page, 292 example values in the spec and every error code explained (15). Semantic API versions and a dated changelog with a pull request behind every line (15). The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema): APIs and MCP servers. - 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool). - 10, llms.txt or Markdown docs served for agents. - 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. - 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs. - 0 to 15, examples and documented error responses. - 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. ## 7. Maintenance & community, 80 out of 100, up to 1.8 more on the total Why it scored 80: Version 2.10.2 on 23 September 2026 (30). Four releases in the last 90 days, 2.9.7 on 9 July and 2.10.0, 2.10.1 and 2.10.2 between 18 and 23 September (20). 32 open issues in the public tracker, labelled and triaged, most of them feature or new-source requests. Two portal bugs carry 'awaiting developer response' since May, and a new-account key error (#656) has been open since 7 July (20 of 25). The hosted MCP server isn't in the official registry, where only a third-party Pipeworx wrapper appears, and there's no official SDK (0). Dependabot, CodeQL and a pytest workflow on the code repository, with dependency bumps merged on 28 September (10). The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance): - 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. - 20, at least three releases or dated changelog entries in the last 90 days. - 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. - 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models). - 10, package health, current dependencies and CI. Models are read for deprecation notice periods and model churn rather than release counts. ## What we couldn't check What 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. - unchecked: status page component history, since pirateweather.xitoring.io renders only with JavaScript - unchecked: prices and limits of plans above $2, behind the JavaScript-only Apiable portal - Whether keys can be rotated or scoped in the Apiable portal isn't documented - The 2025-12-24 outage behind the previous -2 is an availability incident, not one of the negative categories, so we dropped the deduction ## Weaknesses - No SLA, and the terms call it best effort - Key in the URL path by default, and only in the query string for the MCP server - Plans above $2 a month sit behind a JavaScript-only portal, and signup needs a person - No security.txt, SECURITY.md or MCP registry entry - The privacy notice and terms are generator templates from 2024 and say nothing on data reuse ## What costs an agent a turn today The notes we give agents before they call it. Each one is a workaround an agent shouldn't need. - Send the key in an apikey header and put a placeholder in the path so it stays out of logs - Pass blocks= and hourly_indices= or daily_indices= to get only the rows you need. The default payload is large - Ask for version=2 to get fields Dark Sky never had, such as smoke, fire index and the source list - Point historical requests at timemachine.pirateweather.net with a UNIX timestamp after the coordinate - Watch Ratelimit-Remaining. The cap is monthly, so a runaway loop can spend it in a day ## What the review panel asked for - Fix new-account key errors - Issue keys instantly - state data reuse terms ## When it's done Send 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.