Loading ToolBench
MCP server for querying weather information for given cities
Single tool with minimal definition quality. Tool name 'query-weather' lacks action verb clarity (should be 'get_weather' or 'fetch_weather'). Description is only 28 characters, below the 34-char p10 baseline and insufficient for LLM selection context. Input schema present but parameter description 'city name' is generic and lacks format/constraint guidance. No output schema documented. No error recovery guidance. Tool does not follow verb_noun naming convention (pattern:tool). Missing pagination, result limits, and structured output documentation.
query weather for given city
Tool name 'query-weather' lacks action verb. Should be 'get_weather' or 'fetch_weather' to signal intent clearly to LLMs.
Tool description 'query weather for given city' is only 28 characters, below p10 baseline of 34 chars. Insufficient context for LLM selection. Should explain WHEN to use it, WHAT it returns, and any prerequisites.
No output schema documented. LLMs cannot plan downstream calls or extract fields from the JSON response. Must document return structure (e.g., temperature, humidity, condition fields).
Parameter 'city' description 'city name' is generic. Missing format guidance, examples of valid input, and handling of multi-word cities or ambiguous names. Should state: 'City name (e.g., New York, London). Multi-word cities supported.'
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |
Error handling returns raw error messages without recovery guidance. 'query weather failed: <error>' tells LLM nothing actionable. Should categorize errors (retryable vs. user-fixable) and suggest next steps.