get_security_hotspots

Find high-risk security surfaces: authentication code, user input handling, SQL operations, and other OWASP-relevant patterns.

Parameters

Parameter	Type	Default	Description
severity	str	"medium"	Minimum severity: "low", "medium", "high", "critical"
pattern_set	list	all	Filter to: "auth", "input", "sql", "crypto", "file_io"
file_path	str	null	Limit to a specific file

Response

{
  "hotspots": [
    {
      "file_path": "src/auth/login.py",
      "pattern": "auth",
      "severity": "high",
      "symbols": ["verify_password", "create_session"],
      "risk_score": 8.4,
      "reason": "Authentication logic in high-churn file with bus factor 1"
    },
    {
      "file_path": "src/api/search.py",
      "pattern": "input",
      "severity": "medium",
      "symbols": ["search_handler"],
      "risk_score": 5.2,
      "reason": "User input handler with no test coverage"
    }
  ]
}

Pattern Detection

Patterns are detected by combining: - Symbol name heuristics (login, password, token, execute, query, etc.) - Import analysis (hashlib, sqlite3, subprocess, etc.) - Hotspot score (high-churn risky code = higher priority)