changelog-generator
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (HIGH): Vulnerable to Indirect Prompt Injection via git commit messages.\n
- Ingestion points: Git commit history and user-provided style guide files (e.g.,
CHANGELOG_STYLE.md).\n - Boundary markers: Absent; no delimiting or instruction-shielding is implemented.\n
- Capability inventory: File writing (saving results to
CHANGELOG.md) and git command execution.\n - Sanitization: Absent; commit content is treated as trusted data.
Recommendations
- AI detected serious security threats
Audit Metadata