conventional-commits

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run a bundled Node.js script (scripts/check-commit-message.mjs) to automate the validation of Git commit headers.
  • [DATA_EXFILTRATION]: The validator script reads local files provided as command-line arguments to check their content against the Conventional Commits rules. This access is restricted to reading local file content for validation and does not involve network activity.
  • [PROMPT_INJECTION]: The skill processes external commit message text, which creates an indirect prompt injection surface. The validator script mitigates this risk by using regular expressions to enforce a strict schema without interpreting the data as instructions.
  • Ingestion points: The validator reads from file paths or standard input via the readMessage function in scripts/check-commit-message.mjs.
  • Boundary markers: No specific delimiters or boundary warnings are present in the validation instructions.
  • Capability inventory: The skill uses local shell execution to run the validation script; no other high-privilege capabilities are used.
  • Sanitization: The script filters Git comment lines and validates the header using a predefined regular expression.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 10:18 PM
Security Audit — agent-trust-hub — conventional-commits