retro
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes session logs and repository primary sources which can contain untrusted data, creating an attack surface for indirect prompt injection.
- Ingestion points: Session logs and repository files (Step 2,
SKILL.md). - Boundary markers: Absent; there are no instructions to differentiate log data from instructions.
- Capability inventory: The skill suggests modifications to
AGENTS.md,CODING_STANDARDS.md, and CI/linter configurations. - Sanitization: No sanitization or validation of the ingested data is defined.
- [DATA_EXFILTRATION]: The skill accesses local session logs which may contain sensitive development information. It also explicitly suggests expanding agent access to dev server logs and third-party services, which increases the data exposure surface of the environment.
Audit Metadata