cleanup
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill processes uncommitted code changes, creating a surface for indirect prompt injection. If the code being cleaned contains malicious instructions (e.g., in comments), the agent could be influenced. Evidence: 1. Ingestion points: Local git changes (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Relies on host agent file access. 4. Sanitization: Absent.
- [General Security] (SAFE): No malicious code, credential exfiltration, or remote execution patterns were detected. The skill is purely instructional and contains no executable content.
Audit Metadata