plan-secrets-audit
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by scanning repository files and git history, which represents a surface area where instructions embedded in that data could attempt to influence the AI's behavior.\n
- Ingestion points: Repository working tree files and git commit history.\n
- Boundary markers: The skill includes 'Guardrails' and 'Self-critique' sections that enforce a 'plan-only' mode and prevent autonomous execution.\n
- Capability inventory: File content scanning, credential classification, and report generation in markdown format.\n
- Sanitization: Instructions strictly limit the output of any detected sensitive values to the last four characters only.
Audit Metadata