Gets the latest documentation and code examples for any library.
Single tool 'get-library-docs' has a well-structured schema with proper Zod validation and comprehensive description (450+ chars). However, the description is verbose and includes implementation details (token preprocessing) that belong in docs, not the LLM-facing description. Parameter descriptions are present but could be more concise. The tool lacks error handling guidance, output schema documentation, and does not declare whether it is idempotent or read-only (though risk=READ_ONLY is noted in metadata). No tool annotations (readOnlyHint) in the actual schema. The description mixes usage instructions with tool purpose, which is acceptable but not optimal for LLM parsing.
Retrieves up-to-date documentation and code examples for any library. This tool automatically searches for the library by name and fetches its documentation. Usage: 1. Provide the author and library name pair (e.g., "vercel/next.js", "shadcn-ui/ui", "vuejs/docs") 2. Specify a topic to focus the documentation on (e.g., "dynamic", "routing", "authentication") The tool will: 1. Automatically find and select the most relevant library based on the provided name 2. Fetch comprehensive documentation for the selected library 3. Return relevant sections focused on the specified topic Response includes: - Library selection explanation - Comprehensive documentation with code examples - Focused content if a topic was specified
Tool description is verbose (450+ chars) and includes implementation details (token preprocessing, default value logic). LLM-facing descriptions should be 50 - 200 chars, focused on WHAT and WHEN, not HOW. Current description buries the core action in usage instructions.
No output schema documented. LLMs cannot plan downstream calls or extract fields without knowing the response structure. Must document: what fields are returned, their types, and which are guaranteed vs optional.
No error handling guidance. Tool description does not explain what happens on failure (library not found, API timeout, invalid topic). LLMs need recovery hints: 'If library not found, try a different author/name pair.'
Tool annotations missing from schema. Although metadata declares risk=READ_ONLY, the actual Zod schema lacks readOnlyHint annotation. This prevents clients from rendering the tool correctly or applying safety policies.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 63 | <=2025-11-25 | v2 |
Parameter 'tokens' has a complex transformation (preprocess, transform, optional) that is implementation-specific. The description should state the constraint plainly: 'Must be >= 10000. If lower, will be clamped to 10000.' Avoid exposing Zod logic in descriptions.