CodeSage provides 6 code-analysis tools with basic descriptions and input schemas. All tools have verb-noun naming (search, find_symbol, find_references, list_dependencies, impact_analysis, export_context) and documented parameters with types. However, descriptions are minimal (10-50 chars), parameter descriptions lack detail on constraints/formats, and output schemas are not documented. No error handling guidance, no examples of recovery paths, and no indication of pagination support for potentially large result sets. Tools are read-only (low risk), but lack the depth expected for production use.

• Expand tool descriptions to 150-250 chars. Example: 'search: Find code by semantic similarity or keyword matching. Use semantic search for intent-based queries ("parse JSON"), keyword search for exact terms. Returns up to 20 results with file paths and line numbers. Call this first to locate relevant code before analyzing dependencies.'
• Add format/constraint details to parameter descriptions. Example: 'query (string, 1-500 chars): Search term. Semantic queries work best with 3-10 words; keyword queries accept regex patterns.'
• Document output schemas for each tool. Example: 'search returns: {results: [{file: string, line: number, symbol: string, score: float}], total: integer, has_more: boolean}'
• Add error handling guidance. Example: 'If symbol not found, try search() with a partial name or broader scope. If file path is invalid, use list_dependencies() to discover valid files.'
• Add limit and offset parameters to search() and find_references() with documented defaults (e.g., limit=20, offset=0) and caps (max 100).

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry