Single tool 'rugradar_scan' has a reasonable description (127 chars, within 10-1024 range) but critical schema gaps. Input schema present but minimal: only 'target' parameter with basic string type and description. No output schema documented. No error handling guidance, no parameter constraints (format, length, allowed values), no enum for severity levels in findings. Tool name lacks action verb prefix (should be 'scan_contract' or 'analyze_contract'). No pagination support despite potentially returning large finding lists. Descriptions lack WHEN/WHY context for LLM selection.

• Rename tool to 'scan_contract' or 'analyze_contract_risk' to start with action verb. Update description to: 'Scans Solidity contract source or ABI for security risks: honeypots, hidden mint/blacklist, owner backdoors, unlocked liquidity. Returns structured findings with severity levels.'
• Document output schema in tool description: 'Returns {tool: string, findings: [{id: string, sev: enum(high|medium|low), where: string}], score: number}.'
• Add parameter constraints to 'target': 'Path to .sol file, ABI JSON, or directory (max 1000 chars, no symlinks). Relative or absolute paths accepted. Required.'
• Add error handling guidance: 'If target not found or unreadable, returns error with path and reason. If ABI is malformed, returns error with line number. Retry with corrected path.'
• Add pagination support: Accept optional 'limit' parameter (1-100, default 50) and return 'total_findings' count in response to enable multi-call scanning of large codebases.