MCP server that provides access to Yu-Gi-Oh! card information via the YGOProDeck API
Single tool 'yugioh-api' has a comprehensive JSON Schema with 24 parameters, all typed and constrained with enums/ranges. Description is adequate (44 chars) but minimal. Parameter descriptions are present and specific (e.g., 'Filter by exact card name'). However, output schema is completely undocumented, the tool returns raw API response body as a string with no guidance on structure, fields, or how to parse results. No error handling guidance. Tool name lacks action verb (should be 'search_cards' or 'get_cards'). No pagination guidance despite potentially large result sets.
Get information about yugioh cards
Tool name 'yugioh-api' lacks action verb. Should be 'search_cards' or 'get_cards' to signal intent clearly to LLM.
Output schema completely undocumented. Tool returns raw string response body with no guidance on JSON structure, field names, or how to extract card data. LLM cannot plan downstream operations.
No pagination support documented. Tool accepts 'num' and 'offset' parameters but no guidance on result limits, total count, or when to paginate. Large result sets risk context window exhaustion.
No error handling guidance. Code catches exceptions and returns 'Error: <message>' as plain text. LLM cannot determine if error is retryable, user-fixable, or fatal.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 44 | <=2025-11-25 | v2 |
Tool description too generic (44 chars). Does not explain when to use this tool vs alternatives, what data it returns, or prerequisites. Baseline for A+ tools is 50-200 chars with clear intent.