Three tools with complete schemas and reasonable descriptions, but significant gaps in parameter documentation and output schema clarity. All tools are read-only (good for safety), but descriptions lack actionable guidance for LLM selection. Parameter descriptions are present but minimal (averaging ~30 chars vs. baseline 72 chars). No output schemas documented. Error handling exists but generic. Naming is clear and verb-based (query, context, impact).

• Document output schema for each tool. E.g., 'query returns {symbols: [{uid, name, kind, confidence, file, line}], total, nextCursor}'. Include field types and when each is present.
• Expand parameter descriptions to 50-100 chars with format hints. E.g., 'uid: Unique symbol identifier (string, e.g. 'src/main.ts:MyClass'). Use if you have it; otherwise provide name + repo.'
• Add mutual exclusivity rules: 'Provide uid OR name (not both). If neither, repo defaults to most recently indexed.' State this in each tool description.
• Enhance error messages with recovery paths. E.g., 'Symbol not found. Try query() to search by name, or check repo parameter matches available repos: [list].'
• Add result limits and pagination to query tool. Cap default limit at 20, max 500. Return total count and nextCursor for large result sets.