simplify
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data including source code, caller logic, and test suites. This creates an inherent surface for indirect prompt injection where malicious instructions could be embedded in the code or comments being processed. The skill mitigates this risk by instructing the agent to strictly preserve existing behavior, verify all changes through testing, and maintain ownership boundaries.
- Ingestion points: Repository guidance, source code, test files, and worktree diffs (SKILL.md, Step 1).
- Boundary markers: None explicitly defined within the skill instructions.
- Capability inventory: File modification and execution of testing, linting, and type-checking tools (SKILL.md, Step 4).
- Sanitization: Manual diff review and automated verification through the repository's test suite.
Audit Metadata