Reusable MCP server for generating, validating, and managing GitHub issues with AI-powered duplicate detection
Critical gaps across all dimensions. Tool definitions are inferred from examples (stdio-client/index.mjs) rather than visible in source code, no explicit schema registration found in provided code. Descriptions are present but generic (10-60 chars, well below 50-200 char baseline). No input schemas visible in source. Parameters undocumented. No output schema documentation. Error handling absent. This server fails to meet production-grade tool definition standards.
Check for duplicate issues using AI-powered similarity detection
Create a new GitHub issue in the specified repository
Generate a structured GitHub issue from provided details
List all open GitHub issues in a repository
List all available labels in a GitHub repository
No input schemas visible in source code. Tool definitions inferred from examples only, not from explicit registration with JSON Schema.
Descriptions are too short (10-35 chars) and lack context. Baseline for A+ tools is 50-200 chars. Missing WHAT the tool does, WHEN to use it, and prerequisites. Example: 'List all open GitHub issues in a repository' lacks pagination info, return structure, and when to call it vs other tools.
No parameter descriptions visible. LLMs cannot infer what 'title', 'body', 'labels' mean without explicit descriptions.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |
No output schema documentation. LLMs cannot plan downstream calls or extract required fields (e.g., issue_id for chaining). Baseline: 100% of A+ tools document return types.
No error handling guidance. Tools that call external APIs (GitHub, Gemini) will fail, but no recovery hints provided. Missing pattern: error responses should tell LLM what to do next.