Loading ToolBench
A CLI chat application that integrates MCP servers with Claude for document management and interaction
Two tools with basic schemas and descriptions, but significant quality gaps. read_doc_content has a clear, concise description (59 chars) and proper schema with typed parameter. edit_document has a typo in description ('Edir' instead of 'Edit'), acceptable schema, but both tools lack output schema documentation. Parameter descriptions are present but minimal. No error handling guidance, no idempotency hints for the write operation, and no recovery paths. The edit_document tool is destructive but lacks confirmation/dry-run patterns. Overall, definitions are functional but fall short of production quality.
Edir a document by replacing a string in the documents content with a new string
Read the contents of a document and return it as a string.
edit_document description contains typo ('Edir' instead of 'Edit') and lacks destructive operation warning. No mention that this modifies state or guidance on when to use it.
No output schemas documented for either tool. LLMs cannot plan downstream operations or extract structured data without knowing return types.
edit_document is a destructive write operation with no confirmation, dry-run, or undo capability. Agents can corrupt documents without safeguards.
Error handling is minimal. ValueError exceptions are raised but no recovery guidance provided. LLMs receive raw errors with no actionable next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 47 | 2026-07-28+ | v2 |
Parameter descriptions are terse (e.g., 'Id of the document to read'). No format constraints, no guidance on valid doc_id values, no examples of available documents.