git-conventional-commits

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or dangerous capabilities were identified. The skill is limited to providing text formatting instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted git diffs provided by users, creating a potential surface for prompt injection. 1. Ingestion points: Git diff content in the prompt (ref: evals/evals.json). 2. Boundary markers: Evaluation prompts use triple backticks for diffs. 3. Capability inventory: Restricted to text generation; no tools or command execution are allowed in the frontmatter. 4. Sanitization: No explicit sanitization mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 09:25 PM
Security Audit — agent-trust-hub — git-conventional-commits