Loading ToolBench
MCP server that queries weather information for Brazilian cities
The 'add' tool is trivial and also lacks description. Neither tool has documented output schemas. Parameter descriptions exist in queryWeatherByCity but are minimal (10-20 chars each). No error handling guidance, no recovery paths, no output structure documentation. Naming is reasonable (verb_noun pattern) but descriptions are the critical blocker.
Missing tool descriptions entirely. Both queryWeatherByCity and add lack any description field in tool registration. LLMs cannot determine when or why to select these tools without descriptions.
No documented output schema. Tools return structured responses but LLMs have no schema to understand what fields to expect. queryWeatherByCity returns weather data; add returns a sum, neither is documented.
Parameter descriptions are too brief (10-20 chars). 'City name', 'Start date', 'End date' lack context on format, constraints, or when to use them. Baseline is 72 chars average.
No error handling or recovery guidance. If a city is not found or date parsing fails, the tool returns plain text. No actionable error messages to guide LLM retry or fallback.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 39 | 2026-07-28+ | v2 |
'add' tool is generic and out of scope for a weather server. Naming is weak ('add' is not verb_noun; 'calculate_sum' would be clearer). This tool appears to be a placeholder and should be removed or renamed.