Five tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (semgrep_scan, semgrep_findings, security_check), which is good. However, descriptions are generic (avg ~60 chars, below the 194-char baseline), and critical parameter details are missing. No output schemas documented. semgrep_scan, semgrep_scan_local_files, and semgrep_scan_rpc are near-duplicates with unclear distinctions. Parameters like 'config' lack enum constraints despite accepting specific values ('p/docker', 'p/xss', 'auto'). Error handling and recovery guidance absent.

• Consolidate semgrep_scan, semgrep_scan_local_files, and semgrep_scan_rpc into a single tool with a 'source' parameter (enum: 'code', 'local_files', 'rpc') or split with crystal-clear distinctions in descriptions explaining when each is appropriate.
• Convert 'config' parameter to an enum: ['p/docker', 'p/xss', 'p/owasp', 'auto', 'custom'] with description: 'Semgrep ruleset profile. Use p/docker for container security, p/xss for web vulnerabilities, auto for all rules, or custom for user-defined rules.'
• Document output schemas for all tools. Example for semgrep_scan: {findings: [{rule_id, message, severity, file, line, column}], total_issues: int, scan_time_ms: int}.
• Expand tool descriptions to 100 - 200 chars. Example: 'Scan code files for security vulnerabilities using Semgrep rules. Returns findings with severity, location, and remediation guidance. Use for pre-commit checks or CI/CD integration.'
• Add error handling guidance: 'If scan fails, check that code_files are valid JSON with filename and content keys. For RPC scans, verify Semgrep Pro Engine is installed and SEMGREP_APP_TOKEN is set.'