Single tool with adequate naming and basic schema, but limited parameter documentation and no output schema definition. Tool name 'get_weather' follows verb_noun convention. Input parameters have types (number) and basic descriptions, but lack constraints (min/max bounds, precision hints). No documented output schema, LLM cannot predict result structure. Error handling not visible in source. Overall: fair foundation, but gaps prevent confident production use.
Get the weather for a given latitude and longitude.
No output schema documented. LLM cannot predict what fields the API returns (temperature, wind_speed, etc.), forcing it to guess how to use the result.
Tool description is minimal (~50 chars). Does not explain WHEN to use it, what the response contains, or any limitations (e.g., requires valid lat/lon, may fail if API is down).
Parameter descriptions lack constraints. 'Latitude coordinate' and 'Longitude coordinate' do not specify valid ranges (-90 to +90 for lat, -180 to +180 for lon), precision, or format expectations.
No error handling guidance visible. If the API returns 400 (invalid coordinates), 429 (rate limit), or 500 (service down), the source code does not show how errors are caught, classified, or communicated to the LLM.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 51 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 49 | - | v1 |
Tool is read-only but lacks an explicit readOnlyHint annotation in the schema. Modern MCP tools should carry toolAnnotations (readOnlyHint, destructiveHint, idempotentHint) in the tool definition.