retro
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes session logs which serve as an ingestion point for potentially untrusted data.
- Ingestion points: The skill searches through and reads session logs as defined in Step 2 of SKILL.md.
- Boundary markers: No specific delimiters or instructions to disregard instructions embedded within logs are specified.
- Capability inventory: The skill suggests modifications to steering files including AGENTS.md and CODING_STANDARDS.md, and repository configuration files like package.json.
- Sanitization: No explicit validation, filtering, or sanitization of the log content is performed.
- Purpose assessment: This ingestion is necessary for the skill's primary function of retrospective analysis and the skill is limited to making suggestions to the user.
- [SAFE]: No other security findings were identified in the skill instructions or metadata.
Audit Metadata