An MCP server that provides weather forecasts and alerts using the National Weather Service (NWS) API
Two tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Both tools have action-verb names (get-alerts, get-forecast) and non-empty descriptions (40-50 chars each). Input schemas use Zod with type constraints (string length, number min/max), but descriptions lack actionable detail about expected formats, error cases, and recovery paths. Output is unstructured text only, no documented response schema. Error handling returns plain text messages without guidance on retryability or next steps. No tool annotations (readOnlyHint, etc.) despite both being read-only operations.
Get weather alerts for a state
Get weather forecast for a location
Output schema not documented. Tools return unstructured text wrapped in {type: 'text', text: '...'} but no schema definition tells LLMs what fields to expect or how to parse results for downstream tool calls.
Parameter descriptions lack actionable constraints. 'Two-letter state code (e.g. CA, NY)' includes an example value that LLMs may reuse literally. Should use enum or pattern constraint instead. Latitude/longitude descriptions omit units (degrees) and coordinate system (WGS84).
Error responses provide no recovery guidance. 'Failed to retrieve alerts data for state: X' and 'This location may not be supported by the NWS API' are informational but do not tell the LLM whether to retry, ask the user, or try a different tool.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 62 | 2026-07-28+ | v2 |
Tool annotations missing. Both tools are read-only (no side effects) but lack readOnlyHint annotation. This prevents LLMs from optimizing caching and retry logic.
No pagination support. If alerts or forecast periods exceed a reasonable limit, the entire result is returned, risking context window exhaustion. No limit parameter or documentation of max results.