Loading ToolBench
Universal catalog of MCP (Model Context Protocol) servers with documentation, health monitoring, and Claude Code integration
Catalogue MCP exposes 2 tools with critical definition gaps. Both tools have descriptions present but significantly undersized (18-61 chars vs. baseline 194 chars average for production tools). Input schemas are present and properly typed, but parameter descriptions are either missing entirely or minimal. Tool naming follows verb_noun convention (resolve_, get_) which is correct, but the descriptions lack the depth needed for reliable LLM tool selection. No output schemas are documented. Error handling guidance is absent. The server implements HTTP transport (Next.js-based), which satisfies transport requirements, but the tool definitions themselves would not pass production code review.
Fetches up-to-date documentation for a library
Resolves a package name to a Context7-compatible library ID
Tool descriptions critically undersized. resolve-library-id: 18 chars; get-library-docs: 61 chars. Baseline production tools average 194 chars. Descriptions lack WHEN to use, prerequisites, and return value guidance. LLMs cannot reliably select between these tools or understand their place in a workflow.
Input parameter 'libraryName' (resolve-library-id) lacks description. Parameter 'tokens' (get-library-docs) lacks description. JSON Schema lacks parameter-level descriptions in properties. LLMs cannot infer that 'tokens' is a budget or limit vs. a list of token IDs.
Output schemas not documented. LLMs do not know what fields resolve-library-id returns (is it 'library_id', 'id', or a full object?) or what structure get-library-docs produces. Baseline: 100% of A+ tools have documented return types.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 44 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 36 | - | v1 |
No error handling guidance. Descriptions do not state what errors can occur (e.g., library not found, invalid token count) or what the LLM should do next (retry, ask user, use fallback). Baseline pattern: error responses should guide recovery.
Parameter 'context7CompatibleLibraryID' (get-library-docs) is required but its format and structure are not described. Is it a UUID, a string key, or a structured object? Parameter 'tokens' lacks min/max bounds or unit specification (tokens as in LLM tokens, authentication tokens, or count?).
Parameter 'topic' (get-library-docs) is optional but its purpose and valid values are not documented. Does it filter docs by topic? Is it required for specific libraries? Undocumented dependencies cause misuse.