Single tool 'dirsight_scan' has a basic description (89 chars) but critically lacks parameter descriptions and output schema documentation. The input schema defines only 'target' as a string with minimal context. No error handling guidance, no output structure documented, and no indication of what 'ranked endpoints' means or what the JSON response contains. The tool name lacks a clear action verb (scan is weak; 'analyze_discovery_output' would be stronger).

• Add explicit output schema to tool description: 'Returns {"findings": [{"id": string, "severity": string, "path": string, "score": number}], "total": number, "high_interest_count": number}' or equivalent.
• Expand 'target' parameter description: 'Path to a ffuf JSON output file or gobuster text output file. Must be readable and contain valid dirbusting results.'
• Clarify ranking criteria in tool description: 'Ranks endpoints by signal-to-noise ratio using heuristics: HTTP status codes, response length anomalies, and keyword patterns. Returns findings sorted by score (highest first).'
• Add error handling examples: 'If file not found, returns error: "Target file not found at <path>. Ensure the path is correct and readable." If format invalid, returns: "Could not parse <format>. Expected ffuf JSON or gobuster text output."'
• Rename tool to 'analyze_discovery_output' or 'rank_web_endpoints' to better signal intent and distinguish from generic 'scan' operations.