Single tool 'search_docs' has a basic description (54 chars) and partial input schema. Description lacks WHEN to use it, prerequisites, or return format guidance. Input schema defines 'query' and 'workspace' with types and brief descriptions, but no output schema is documented. No error handling guidance (e.g., what if docs/ doesn't exist?). Tool name is verb-based (search_) which is correct, but the implementation snippet shows the tool is incomplete, the code cuts off mid-function. No evidence of parameter validation, enum constraints, or structured output documentation. Baseline: A+ tools average 194 chars in description and 100% have documented return types; this tool has neither.
Search project documentation files for relevant content. Scans README, docs/, and markdown files for matches.
No output schema documented. LLMs cannot plan downstream calls or extract structured data from search results.
Description is 54 characters, below the 10 - 1024 baseline and lacks WHEN to use, prerequisites, or return format. Does not answer: What does it return? How many results? What format?
No error handling guidance. What happens if workspace path is invalid? If docs/ doesn't exist? If query matches nothing? LLM has no recovery path.
Parameter 'workspace' lacks constraints. No validation that it's a valid directory path. No guidance on absolute vs relative paths. LLMs may pass invalid paths.
Source code snippet is incomplete (cuts off mid-function). Cannot verify full implementation, error handling, or actual return structure.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 45 | 2026-07-28+ | v2 |