A multi-component system integrating MCP servers for travel planning, weather forecasts, and Google Maps routing with AI agents powered by Groq/LLaMA
This MCP server has critical definition quality gaps across all three tools. Tool descriptions are present but minimal (10-70 chars, below the 50-200 char baseline for LLM-optimized descriptions). Input schemas are properly structured with types and parameter descriptions, which is a strength. However, output schemas are completely undocumented, there is no specification of what these tools return, forcing LLMs to guess at response structure. Error handling is absent; there is no guidance on retryability, recovery steps, or actionable error messages. Tool naming follows verb_noun convention (get_alerts, get_forecast, calculate_route) which is good, but descriptions are too terse to guide LLM selection. The server lacks any mention of pagination for potentially large result sets (e.g., alerts could return hundreds of entries). No idempotency hints, permission scopes, or rate-limit guidance are present. The three tools are READ_ONLY, which is appropriate, but this is not declared in tool annotations (toolAnnotations feature is false).
Calculate route between two geographic coordinates using Google Maps API
Get weather alerts for a US state.
Get weather forecast for a location.
Output schemas completely undocumented. No documentation of what get_alerts, get_forecast, or calculate_route return. LLMs cannot plan downstream operations or extract required fields.
Tool descriptions are too terse (10-70 chars). Baseline is 50-200 chars for LLM-optimized descriptions. Current descriptions lack context for WHEN to use each tool, WHAT format to expect, and any prerequisites.
No error handling or recovery guidance. No indication of what errors are retryable, what are user-fixable, or what are fatal. No actionable error messages documented.
No pagination support documented. Tools may return large result sets (e.g., 100+ weather alerts for a state) with no limit, offset, or cursor parameters. This risks context window exhaustion.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 44 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 38 | - | v1 |
Tool annotations disabled (toolAnnotations=false). Tools are READ_ONLY but this is not signaled via readOnlyHint annotation, forcing LLMs to infer safety from descriptions.
No permission scope declarations. Tools do not state what permissions they require (e.g., 'read:weather', 'write:none'). This prevents least-privilege agent configuration.