MCP server for quillopy - a documentation search tool for programming languages, libraries, and frameworks
Two tools with basic schemas and descriptions, but significant gaps in parameter documentation and error guidance. quillopy_search has 4 parameters with types and descriptions, but lacks enums for constrained inputs (language, documentation_name). quillopy_chat_with_code has only 2 parameters, both described. Neither tool documents output schema structure or pagination. Error handling returns generic text responses without recovery guidance. No tool annotations (readOnlyHint, idempotentHint). Descriptions are adequate (100-150 chars) but lack WHEN/WHY context. Parameter descriptions are present but lack format constraints, ranges, or examples of valid values.
This tool enables chat with code in a GitHub repository. Only use when explicitly requested by the user. Inform the user that responses may take up to 1 minute to generate.
This MCP searches and fetches documentation for programming libraries, packages, and frameworks. When a user types @quillopy or @quillopy[documentation_name], they are requesting to use this tool to access programming documentation.
No output schema documented. Tools return unstructured text responses without field definitions. LLMs cannot plan downstream calls or extract structured data.
Parameters lack format constraints and enums. 'language' and 'documentation_name' accept free-form strings with no validation. Should use enums or regex patterns to prevent hallucinated values.
Error handling returns generic text without recovery guidance. Errors like 'Failed to retrieve documentation: No response received' do not tell the LLM what to do next or whether to retry.
Tool names lack clear action verbs. 'quillopy_search' is acceptable, but 'quillopy_chat_with_code' is vague, does it search, analyze, or execute? Rename to 'search_github_code' or 'analyze_github_repository'.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2026-07-28+ | v2 |
No tool annotations. Tools lack readOnlyHint, destructiveHint, or idempotentHint. Agents cannot determine which tools are safe to retry or have side effects.