MCP server: convert JSON arrays of objects to GitHub-flavored Markdown tables.
Single tool 'render' has a clear, concise description (95 chars) and explicit input schema with proper JSON Schema structure. However, critical gaps limit production readiness: (1) no output schema documented, LLMs cannot predict return structure; (2) parameter descriptions are minimal (align param lacks detail on valid values); (3) no error handling guidance, what happens with invalid rows or malformed objects?; (4) no parameter validation rules stated (e.g., max rows, max columns). The tool itself is well-named (verb_noun pattern) and the implementation is sound, but the definition lacks the rigor expected for A-grade tools.
Render an array of JSON objects as a GitHub-flavored Markdown table. Pipe-escapes cells.
No output schema documented. LLMs cannot predict the structure of the returned markdown table or plan downstream operations.
Parameter 'align' description is vague. States 'Map of column name → "left" | "center" | "right"' but does not explain what happens if an invalid alignment is provided or if a column in align does not exist in columns.
No error handling guidance. Tool catches errors and returns 'mdtable failed: <message>' but does not guide LLM on recovery (e.g., 'Ensure rows is a non-empty array of objects').
Parameter 'rows' lacks validation constraints. No mention of max array size, max object depth, or handling of null/undefined values (though code does handle them).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2026-07-28+ | v2 |
Parameter 'columns' description does not explain behavior when columns is omitted (union of all keys) or when a column in the array does not exist in any row.