The server defines 3 tools with basic structure but significant quality gaps. All three tools (calculator, search, weather) have descriptions and input schemas present, but descriptions are generic and lack the LLM-optimized depth required for reliable tool selection. Schemas are present but lack parameter constraints (enums, ranges, format specifications). The 'operation' parameter in calculator accepts a free-form string instead of an enum constraint, inviting hallucinated invalid values. The search and weather tools lack pagination/limit controls and do not document result structure or size constraints. Error handling is not visible in the tool definitions. This is typical of a mediocre community MCP server (C-grade territory).
Perform mathematical calculations.
Search the web for information.
Get current weather information for a location.
calculator.operation parameter lacks enum constraint, accepts arbitrary string values instead of enforcing {add, subtract, multiply, divide}. LLMs will hallucinate invalid operations.
search tool lacks pagination parameters (limit, offset/page, cursor) and does not document result count limits or truncation behavior. Returning unlimited results will blow context windows.
All tool descriptions are generic and under 100 characters. They do not address WHEN to use the tool vs. similar tools, what happens on error, or what output structure to expect. LLMs lack signal for tool selection.
No output schemas documented. LLMs cannot predict what fields the tools return, preventing downstream tool chaining and forcing agents to guess response structure.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 52 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 38 | - | v1 |
weather.units parameter accepts free-form string, should be enum [metric, imperial] to prevent LLM errors and invalid API calls.
calculator description does not state that tool is idempotent or safe to retry. search description does not explain when to call it vs. other discovery tools. weather does not document rate limits or cache behavior.
No error handling documentation visible. Tools do not describe failure modes (e.g. what happens if search returns no results, if calculator gets division by zero, if weather API is unavailable). Agents lack recovery guidance.