code-review
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from git diff and git log outputs and interpolates them directly into a subagent prompt.
- Ingestion points: git diff and git log commands in Phase 1 (SKILL.md).
- Boundary markers: Absent. The review prompt template in Phase 2 lacks explicit delimiters or instructions to ignore embedded commands within the diff output.
- Capability inventory: The skill uses shell commands for repository inspection and the Agent tool for subagent dispatch.
- Sanitization: None detected.
- [COMMAND_EXECUTION]: The skill utilizes shell commands to inspect the local repository state.
- Evidence: git diff, git log, and ls docs/plans/*.md are used to gather context for the review.
Audit Metadata