A multi-server MCP project with brand comparison, weather, and Wikipedia search capabilities
Server has 5 tools across multiple files with inconsistent quality. All tools have descriptions (good), but schemas are incomplete or missing type information. Tool names follow verb_noun convention (get_*, search_). However, parameter descriptions are minimal (1-2 words), input schemas lack full type definitions in several cases, and output schemas are undocumented. No error handling guidance, no pagination support for list-like results, and no security considerations visible. The server appears to be a collection of demo/prototype implementations rather than production-grade tools.
Fetches real-time shopping data for a specific brand and product type.
Fetches top 3 product prices for a brand using SerpApi.
Get current weather for a city
Search Wikipedia and return a short summary.
Tool 'greet' has no description. LLMs cannot determine when or why to select it.
Output schemas are not documented for any tool. LLMs cannot plan downstream calls or extract required fields (e.g., brand_id, product_id for chaining).
Parameter descriptions are minimal (1-2 words: 'The brand name', 'The city name'). LLMs need format hints, constraints, and examples of valid values.
No error handling or recovery guidance. Tools return raw API responses without actionable error messages for LLMs (e.g., 'No summary found' is vague; should suggest alternatives).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 59 | 2025-06-18+ | v2 |
API keys (SERP_API_KEY) are loaded via environment variables but no validation or rate-limiting is visible. No audit trail or permission gates for destructive operations.