Single tool 'convert_markdown_to_docx' has a clear, action-verb name (good) and a 54-character description that meets minimum length. Input schema is visible with 4 parameters, all typed and described. However, the description lacks WHEN to use it, prerequisites, or what the tool returns, critical gaps per pattern:tool-description. No output schema documented. The 'config' parameter accepts a file path but lacks validation guidance. No error handling guidance (e.g., what if markdown is malformed or output path is unwritable). Tool is functional but falls short of production-grade LLM-optimized documentation.
Convert a Markdown file to DOCX format.
Tool description lacks WHEN to use it and what it returns. Current: 'Convert a Markdown file to DOCX format.' Should explain: when to call (e.g., 'when user wants to export markdown as a Word document'), what it returns (e.g., 'writes DOCX file to disk; returns success/error status'), and any prerequisites (e.g., 'requires valid markdown input').
No output schema documented. LLMs cannot plan downstream actions or extract results. Must document: does the tool return a file path, file size, success boolean, error message? Current code shows tool writes to disk but response structure is not visible in source.
'config' parameter description says 'Path to a JSON config file (may include preset, style, ignoreImage, math, etc.)' but does not specify format, required fields, or validation rules. LLMs cannot construct valid config files without explicit schema or examples.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | 2026-07-28+ | v2 |
No error handling guidance. Tool description does not explain what happens if: markdown file does not exist, output path is unwritable, markdown is malformed, or preset is invalid. Per pattern:recovery-guide, errors must tell the LLM what to do next.
STDIO transport only. Server is not remotely accessible and cannot be used by hosted MCP clients. Limits deployment to local/CLI-only scenarios.