Four tools with minimal documentation. All have descriptions (10-60 chars, below the 194-char baseline), but lack parameter descriptions entirely. Input schemas present but sparse: explain_rationale has one string param with description; others have minimal or empty schemas. No output schemas documented. No error handling guidance. Tool names follow verb_noun pattern but descriptions are too terse to guide LLM selection. No evidence of idempotency, pagination, or composition patterns.

• Expand tool descriptions to 100-200 chars. Include WHAT the tool does, WHEN to use it (vs similar tools), and what it returns. E.g., 'explain_rationale: Retrieve the recorded rationale (reason, decision, context) for a specific code location or commit. Use this to understand why a change was made. Returns rationale text and metadata.'
• Add descriptions to all input parameters. For 'target', explain the three accepted formats (path:line, path:start-end@revision, commit:revision) and provide a constraint pattern. For 'record_id', explain what a stable record identifier is and how to obtain one.
• Document output schemas for each tool. Specify the fields returned (e.g., rationale_text, author, timestamp, source) and their types. This enables LLMs to chain tools and extract required data.
• Add error handling guidance. E.g., 'If target not found, try search_candidate_evidence() to locate related records. If record_id is invalid, call find_rationale_gaps() to discover available records.'
• For find_rationale_gaps, clarify the input schema. If it takes no parameters, document that explicitly. If it requires context (e.g., working directory), add parameters or explain how context is inferred.