SpiderBrain provides 6 tools with complete JSON schemas and descriptions. However, most descriptions are domain-specific and lack LLM-optimized guidance on WHEN to use each tool vs. alternatives. Parameter descriptions are minimal or absent (e.g., sb_keystones 'n' param has no description). No error handling guidance, no recovery paths, and no output schema documentation. Tools are well-named (verb_noun pattern) but descriptions average ~80 chars, below the 194-char baseline for A+ tools. No tool annotations (readOnlyHint, etc.) despite all being read-only operations.

• Add parameter descriptions to all tools. For sb_keystones, document 'n' as 'Number of top keystones to return (default: 10, range: 1-100)'. For sb_map and sb_path, clarify 'repo-relative file path' format and whether it must exist.
• Document output schemas in tool descriptions. E.g., 'Returns {id, found, reached: [{id, cost}]}' so LLMs know what fields to expect and can chain tools correctly.
• Add readOnlyHint: true to all 6 tool definitions in MCP registration. This signals to agents that these tools are safe to call without confirmation.
• Enhance tool descriptions with WHEN/WHY guidance. E.g., 'Use sb_blast for a single file's impact; use sb_impact for multiple changed files and keystone detection.' This reduces LLM confusion when multiple similar tools exist.
• Add error handling examples to sb_ask description: 'Requires a valid SpiderBrain API key in environment. If missing, returns error with setup instructions.' This guides agents on recovery.