{
  "fixes": {
    "slug": "meteomatics",
    "name": "Meteomatics Weather API",
    "listing": "https://www.anchorterminal.com/tools/meteomatics",
    "markdown": "# Fix list: Meteomatics Weather API\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/meteomatics, the October 2026 research run, assessed 8 October 2026. Grade D, 49.6 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 Meteomatics Weather API: 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. Reliability, 34 out of 100, up to 13.2 more on the total\n\nWhy it scored 34: Graded as a hosted service. No status page was found. The site and legal pages link none, meteomatics.statuspage.io redirects to an inactive notice, and status.meteomatics.com didn't connect for us (0), so there is no incident record to read (0). The docs give per-query limits with numbers, 10 locations and 10 parameters a query, 2,000 characters on a GET and a 5-minute cut-off, and say the query count is set by contract and readable at `/user_stats_json`. No request rate is published (8 of 15). 429 is documented as licence limits reached and 408 comes with advice to shrink the query. No `Retry-After` or backoff guidance was found, and every call is a read (6 of 15). The Service Level Agreement of 7 May 2025 aims for 99 per cent a month on Standard and guarantees 99.5 and 99.9 per cent on Extended and Priority Support, compensated in free months (10). The REST API has been online since May 2015 per the docs. The MCP server is described as beta in the vendor's 4 May 2026 guide (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## 2. Payments \u0026 pricing, 0 out of 100, up to 12.5 more on the total\n\nWhy it scored 0: No x402, MPP or L402 in the docs or the OpenAPI file. The error table lists 402, with no payment protocol behind it (0). The pricing page says prices are based on usage and gives no figure, only a form to talk to sales (0). No free tier or trial is on the site. The general terms let Meteomatics grant a trial through an individual offer (0). A person talks to sales and receives a username and password by email. The MCP server registers OAuth clients by API but still needs that account (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## 3. Security \u0026 auth, 54 out of 100, up to 8.1 more on the total\n\nWhy it scored 54: REST calls use HTTP Basic with a username and password sent by email, or a token from login.meteomatics.com that lasts 2 hours and is documented as an `access_token` query parameter. No key rotation or separate keys were found in the docs. The MCP server uses OAuth with dynamic client registration, the authorisation code grant with PKCE (S256), refresh tokens and one scope, `weather:read` (12 of 30, after the 10-point deduction for a secret in the URL). The whole API is read-only and the single OAuth scope is a read scope (15 of 20). Responses are mostly numbers. Aviation bulletins and official alerts carry third-party text, and the MCP guide tells users to check model output, with no guidance on untrusted content (9 of 15). `/user_stats` and `/user_stats_json` report query counts against contract limits, and the privacy notice says usage logs are kept for 3 months. No per-call log for the customer was found (5 of 15). ISO 27001 certification announced on 23 June 2026 with the certificate published, Cyber Essentials, and a security contact for vulnerability reports. No security.txt (404) and no bug bounty (13 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. Maintenance \u0026 community, 42 out of 100, up to 5.1 more on the total\n\nWhy it scored 42: The newest dated product update is 24 August 2026 (an icing warning parameter and radar sources for Australia, Turkey and Japan), 45 days before this check (20 of 30). That is one dated entry in the last 90 days, not three (0). Product update posts on 26 March, 15 June and 24 August 2026, support by email around the clock per the SLA, and a pull request merged on the Python connector on 15 June 2026 (9 of 15). The Python connector reached 4.0.0 on PyPI on 29 April 2026 and needs Python 3.9 or later. The MCP server isn't in the official registry (10 of 15). The connector repository has no CI workflow, tests or changelog file (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. Schema \u0026 documentation, 71 out of 100, up to 4.7 more on the total\n\nWhy it scored 71: OpenAPI 3.0.0 file at swagger-ui.meteomatics.com/meteomatics_swagger_definition.yaml with 31 paths and 31 operations. The docs say it is documentation only and doesn't list every weather parameter. The MCP server gives JSON Schema inputs for all nine tools (22 of 25). `llms.txt` and `llms-full.txt` at www.meteomatics.com index the site and the API docs (10). MCP tool descriptions say when to use each tool and what not to put in a field, and the docs explain each URL segment and option (15 of 20). The OpenAPI file has 43 enums, but the core request is a URL grammar of free strings, more than 1,800 parameter names aren't enumerated, and MCP inputs such as `parameters`, `aggregation` and `crs` are plain strings with allowed values in prose (8 of 15). Example URLs and responses on every docs page, 149 examples in the OpenAPI file and a table of 17 HTTP codes, with errors returned as plain text (11 of 15). No API version in the path and no API changelog. JSON responses carry `version` 3.0 and product update posts appear every two to three months (5 of 15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):\n\nAPIs and MCP servers.\n\n- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.\n\nModels are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.\n\n## 6. Transparency \u0026 trust, 59 out of 100, up to 3.6 more on the total\n\nMade of editorial 50, provenance 68.\n\nWhy it scored 59: Closed service under general terms for enterprise customers dated August 2026, published per group company, with Swiss law and the courts at the vendor's seat. The Python connector is MIT (15 of 30). The privacy notice, last updated 20 August 2026, names the Weather API, says technical data and usage logs are kept for 3 months, and names Meteomatics AG as controller. A published Data Processing Agreement covers breach notice, deletion or return at the end of the contract and 30 days' notice of a new sub-processor (22 of 30). No deprecation policy was found. The terms let the vendor change them by posting a new version, and a commit marked as breaking on the Python connector, dated 29 April 2026, deprecates the `ecmwf-ens-cluster` source with no notice period stated (3 of 20). The privacy notice names nine services with their countries, among them HubSpot, Stripe, Gong and OpenAI. No hosting provider or data location for the API is named (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- Terms of service: read, states 6 of the 7 things a reader expects (9.1 of 10)\n- Privacy policy: read, states 6 of the 8 things a reader expects (8.5 of 10)\n- Status page: not found (0 of 10)\n- Changelog: not found (0 of 10)\n- security.txt: not found (0 of 10)\n\n## 7. Agent ergonomics, 80 out of 100, up to 3.3 more on the total\n\nWhy it scored 80: The parameter list, time range, step and format in the URL size each response, and the MCP server has nine tools in about 15 KB of definitions (23 of 25). No pagination. Output is bounded by the time range and interval, 10 locations and 10 parameters a query, a `timeout` option and a choice of JSON, CSV, XML, NetCDF or images (13 of 20). HTTP codes with a plain-text message that says which part of the URL is wrong or what data is available, 206 for partial content, and no machine-readable error key (12 of 20). Every REST call is a read, including the POST form, and all nine MCP tools carry `readOnlyHint` true and `destructiveHint` false (20). The default source is the vendor's mix model. A call needs four URL segments and parameter names with units. The Python connector is on PyPI at 4.0.0, and connectors for R, Go, Java, C++, Kotlin and Julia are on GitHub (12 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## What we couldn't check\n\nWhat we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.\n\n- unchecked: whether a status page exists at status.meteomatics.com. The address didn't connect for us, the site links no status page, and meteomatics.statuspage.io redirects to an inactive notice\n- unchecked: authenticated behaviour, including response headers on 429, error bodies and the shape of `/user_stats_json`. We had no credentials and called no data endpoint or MCP tool\n- unchecked: open issues and pull requests on the connector repositories. The GitHub API refused us for its rate limit, so we read the cloned history only\n- Whether a trial or test account is still granted. The terms allow a trial through an individual offer and the Python connector has an example named `12_test_account_options.py`, but the site shows none\n- The MCP server answered `initialize` and `tools/list` without a token on 8 October 2026. The docs page doesn't say it is beta, while the 4 May 2026 guide does\n- The OpenAPI file lists http://api.meteomatics.com as a second server and one docs example uses http. We didn't test whether plain HTTP answers\n- The general terms are published per group company. We used the Meteomatics AG version, dated August 2026, and didn't compare the GmbH, Limited and Inc. versions\n- The lead was right. The MCP entry among the data connectors is a hosted server with OAuth, and credentials are issued after a sales conversation\n\n## Weaknesses\n\n- No public price, free tier or self-serve signup found. Credentials arrive by email after a sales conversation\n- No status page found. meteomatics.statuspage.io redirects to an inactive notice and the site links none\n- REST calls use HTTP Basic with the account username and password, and the two-hour token may travel as an `access_token` query parameter\n- No request rate is published. The query count is set by contract, and 429 carries no documented `Retry-After`\n- No API changelog or deprecation policy found. Product update posts appear every two to three months\n- The MCP server is described as beta in the vendor's 4 May 2026 guide and isn't in the official MCP registry\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- Ask the account owner for credentials first. There is no signup, and both the REST API and the MCP server need a Meteomatics account with API access\n- Build REST URLs as validdatetime/parameters/location/format, for example `2026-10-08T00:00:00Z/t_2m:C/52.52,13.40/json`. Parameter names carry their unit after a colon\n- Keep each query to 10 locations and 10 parameters unless the contract says otherwise, and use POST when a GET URL passes 2,000 characters\n- Send credentials or the token in the `Authorization` header. Avoid the documented `access_token` query parameter, which puts the token in URLs and logs\n- Read `/user_stats_json` before a large job. It reports queries sent and the contract's limits, and 429 means a limit is spent\n- Treat reserved values such as -888 and -999 as invalid or missing data, and shrink the request when a 408 arrives after the 300-second timeout\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": "D",
    "score": 49.6,
    "assessed": "2026-10-08",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "reliability",
        "name": "Reliability",
        "score": 34,
        "maxGain": 13.2,
        "reason": "Graded as a hosted service. No status page was found. The site and legal pages link none, meteomatics.statuspage.io redirects to an inactive notice, and status.meteomatics.com didn't connect for us (0), so there is no incident record to read (0). The docs give per-query limits with numbers, 10 locations and 10 parameters a query, 2,000 characters on a GET and a 5-minute cut-off, and say the query count is set by contract and readable at `/user_stats_json`. No request rate is published (8 of 15). 429 is documented as licence limits reached and 408 comes with advice to shrink the query. No `Retry-After` or backoff guidance was found, and every call is a read (6 of 15). The Service Level Agreement of 7 May 2025 aims for 99 per cent a month on Standard and guarantees 99.5 and 99.9 per cent on Extended and Priority Support, compensated in free months (10). The REST API has been online since May 2015 per the docs. The MCP server is described as beta in the vendor's 4 May 2026 guide (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": "payments",
        "name": "Payments \u0026 pricing",
        "score": 0,
        "maxGain": 12.5,
        "reason": "No x402, MPP or L402 in the docs or the OpenAPI file. The error table lists 402, with no payment protocol behind it (0). The pricing page says prices are based on usage and gives no figure, only a form to talk to sales (0). No free tier or trial is on the site. The general terms let Meteomatics grant a trial through an individual offer (0). A person talks to sales and receives a username and password by email. The MCP server registers OAuth clients by API but still needs that account (0).",
        "checklist": [
          "The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).",
          "- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.\n- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for \"contact sales\" or prices behind a login.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).",
          "Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.",
          "Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-payments"
      },
      {
        "key": "security",
        "name": "Security \u0026 auth",
        "score": 54,
        "maxGain": 8.1,
        "reason": "REST calls use HTTP Basic with a username and password sent by email, or a token from login.meteomatics.com that lasts 2 hours and is documented as an `access_token` query parameter. No key rotation or separate keys were found in the docs. The MCP server uses OAuth with dynamic client registration, the authorisation code grant with PKCE (S256), refresh tokens and one scope, `weather:read` (12 of 30, after the 10-point deduction for a secret in the URL). The whole API is read-only and the single OAuth scope is a read scope (15 of 20). Responses are mostly numbers. Aviation bulletins and official alerts carry third-party text, and the MCP guide tells users to check model output, with no guidance on untrusted content (9 of 15). `/user_stats` and `/user_stats_json` report query counts against contract limits, and the privacy notice says usage logs are kept for 3 months. No per-call log for the customer was found (5 of 15). ISO 27001 certification announced on 23 June 2026 with the certificate published, Cyber Essentials, and a security contact for vulnerability reports. No security.txt (404) and no bug bounty (13 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": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 42,
        "maxGain": 5.1,
        "reason": "The newest dated product update is 24 August 2026 (an icing warning parameter and radar sources for Australia, Turkey and Japan), 45 days before this check (20 of 30). That is one dated entry in the last 90 days, not three (0). Product update posts on 26 March, 15 June and 24 August 2026, support by email around the clock per the SLA, and a pull request merged on the Python connector on 15 June 2026 (9 of 15). The Python connector reached 4.0.0 on PyPI on 29 April 2026 and needs Python 3.9 or later. The MCP server isn't in the official registry (10 of 15). The connector repository has no CI workflow, tests or changelog file (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": "schema",
        "name": "Schema \u0026 documentation",
        "score": 71,
        "maxGain": 4.7,
        "reason": "OpenAPI 3.0.0 file at swagger-ui.meteomatics.com/meteomatics_swagger_definition.yaml with 31 paths and 31 operations. The docs say it is documentation only and doesn't list every weather parameter. The MCP server gives JSON Schema inputs for all nine tools (22 of 25). `llms.txt` and `llms-full.txt` at www.meteomatics.com index the site and the API docs (10). MCP tool descriptions say when to use each tool and what not to put in a field, and the docs explain each URL segment and option (15 of 20). The OpenAPI file has 43 enums, but the core request is a URL grammar of free strings, more than 1,800 parameter names aren't enumerated, and MCP inputs such as `parameters`, `aggregation` and `crs` are plain strings with allowed values in prose (8 of 15). Example URLs and responses on every docs page, 149 examples in the OpenAPI file and a table of 17 HTTP codes, with errors returned as plain text (11 of 15). No API version in the path and no API changelog. JSON responses carry `version` 3.0 and product update posts appear every two to three months (5 of 15).",
        "checklist": [
          "APIs and MCP servers.",
          "- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.",
          "Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-schema"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 59,
        "maxGain": 3.6,
        "reason": "Closed service under general terms for enterprise customers dated August 2026, published per group company, with Swiss law and the courts at the vendor's seat. The Python connector is MIT (15 of 30). The privacy notice, last updated 20 August 2026, names the Weather API, says technical data and usage logs are kept for 3 months, and names Meteomatics AG as controller. A published Data Processing Agreement covers breach notice, deletion or return at the end of the contract and 30 days' notice of a new sub-processor (22 of 30). No deprecation policy was found. The terms let the vendor change them by posting a new version, and a commit marked as breaking on the Python connector, dated 29 April 2026, deprecates the `ecmwf-ens-cluster` source with no notice period stated (3 of 20). The privacy notice names nine services with their countries, among them HubSpot, Stripe, Gong and OpenAI. No hosting provider or data location for the API is named (10 of 20).",
        "blend": "editorial 50, provenance 68",
        "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": "ergonomics",
        "name": "Agent ergonomics",
        "score": 80,
        "maxGain": 3.3,
        "reason": "The parameter list, time range, step and format in the URL size each response, and the MCP server has nine tools in about 15 KB of definitions (23 of 25). No pagination. Output is bounded by the time range and interval, 10 locations and 10 parameters a query, a `timeout` option and a choice of JSON, CSV, XML, NetCDF or images (13 of 20). HTTP codes with a plain-text message that says which part of the URL is wrong or what data is available, 206 for partial content, and no machine-readable error key (12 of 20). Every REST call is a read, including the POST form, and all nine MCP tools carry `readOnlyHint` true and `destructiveHint` false (20). The default source is the vendor's mix model. A call needs four URL segments and parameter names with units. The Python connector is on PyPI at 4.0.0, and connectors for R, Go, Java, C++, Kotlin and Julia are on GitHub (12 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"
      }
    ],
    "provenance": [
      {
        "label": "Terms of service",
        "value": "read, states 6 of the 7 things a reader expects",
        "points": 9.1,
        "max": 10
      },
      {
        "label": "Privacy policy",
        "value": "read, states 6 of the 8 things a reader expects",
        "points": 8.5,
        "max": 10
      },
      {
        "label": "Status page",
        "value": "not found",
        "points": 0,
        "max": 10
      },
      {
        "label": "Changelog",
        "value": "not found",
        "points": 0,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "not found",
        "points": 0,
        "max": 10
      }
    ],
    "unchecked": [
      "unchecked: whether a status page exists at status.meteomatics.com. The address didn't connect for us, the site links no status page, and meteomatics.statuspage.io redirects to an inactive notice",
      "unchecked: authenticated behaviour, including response headers on 429, error bodies and the shape of `/user_stats_json`. We had no credentials and called no data endpoint or MCP tool",
      "unchecked: open issues and pull requests on the connector repositories. The GitHub API refused us for its rate limit, so we read the cloned history only",
      "Whether a trial or test account is still granted. The terms allow a trial through an individual offer and the Python connector has an example named `12_test_account_options.py`, but the site shows none",
      "The MCP server answered `initialize` and `tools/list` without a token on 8 October 2026. The docs page doesn't say it is beta, while the 4 May 2026 guide does",
      "The OpenAPI file lists http://api.meteomatics.com as a second server and one docs example uses http. We didn't test whether plain HTTP answers",
      "The general terms are published per group company. We used the Meteomatics AG version, dated August 2026, and didn't compare the GmbH, Limited and Inc. versions",
      "The lead was right. The MCP entry among the data connectors is a hosted server with OAuth, and credentials are issued after a sales conversation"
    ],
    "weaknesses": [
      "No public price, free tier or self-serve signup found. Credentials arrive by email after a sales conversation",
      "No status page found. meteomatics.statuspage.io redirects to an inactive notice and the site links none",
      "REST calls use HTTP Basic with the account username and password, and the two-hour token may travel as an `access_token` query parameter",
      "No request rate is published. The query count is set by contract, and 429 carries no documented `Retry-After`",
      "No API changelog or deprecation policy found. Product update posts appear every two to three months",
      "The MCP server is described as beta in the vendor's 4 May 2026 guide and isn't in the official MCP registry"
    ],
    "agentNotes": [
      "Ask the account owner for credentials first. There is no signup, and both the REST API and the MCP server need a Meteomatics account with API access",
      "Build REST URLs as validdatetime/parameters/location/format, for example `2026-10-08T00:00:00Z/t_2m:C/52.52,13.40/json`. Parameter names carry their unit after a colon",
      "Keep each query to 10 locations and 10 parameters unless the contract says otherwise, and use POST when a GET URL passes 2,000 characters",
      "Send credentials or the token in the `Authorization` header. Avoid the documented `access_token` query parameter, which puts the token in URLs and logs",
      "Read `/user_stats_json` before a large job. It reports queries sent and the contract's limits, and 429 means a limit is spent",
      "Treat reserved values such as -888 and -999 as invalid or missing data, and shrink the request when a 408 arrives after the 300-second timeout"
    ],
    "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"
  }
}
