Single tool 'ssrfind_scan' has a basic description (67 chars) and minimal schema. Input parameter 'target' lacks type specification in visible code (only inferred as string from decorator). No output schema documented. No error handling guidance, no parameter constraints, no recovery paths. Tool does one thing (scan for SSRF patterns) but lacks production-grade documentation. Naming is clear (verb_noun pattern) but insufficient to overcome schema and description gaps.

• Add explicit JSON Schema for input: { 'type': 'object', 'properties': { 'target': { 'type': 'string', 'description': 'File or directory path to scan for SSRF-prone patterns. Accepts absolute or relative paths. Recursively scans all files.', 'minLength': 1, 'maxLength': 4096 } }, 'required': ['target'] }
• Expand tool description to 150-250 chars: 'Scans source code for SSRF-prone sinks and unvalidated URL fetches. Detects patterns like requests.get(user_input) and urllib.urlopen(f-string). Returns JSON array of findings with severity (high/medium/low), rule ID, and file location. Requires read access to target path.'
• Document output schema: { 'type': 'object', 'properties': { 'tool': { 'type': 'string' }, 'findings': { 'type': 'array', 'items': { 'type': 'object', 'properties': { 'id': { 'type': 'string', 'description': 'Rule ID (e.g., GEN-001)' }, 'sev': { 'type': 'string', 'enum': ['high', 'medium', 'low'] }, 'where': { 'type': 'string', 'description': 'File path where pattern was found' } } } }, 'score': { 'type': 'integer', 'description': 'Total number of findings' } } }
• Add error handling guidance to description: 'Returns empty findings array if target is not found or unreadable. Skips binary files and symlink loops. Timeout: 30s per file.'
• Add parameter description: 'target: Filesystem path (file or directory) to scan. If directory, recursively scans all readable files. Relative paths are resolved from current working directory.'