merge-ready
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of code diffs (
git diff origin/main...HEAD) which are processed by subagents during the review phase. - Ingestion points: Code changes and comments within the branch are evaluated by multiple subagents in Step 2.
- Boundary markers: The instructions do not specify explicit delimiters for the code diffs, though the subagents are given a narrow scope (specific axes like security or billing).
- Capability inventory: The skill utilizes shell commands (
git,pnpm), performs file modifications (automated fixes), and interacts with remote repositories (git push). - Sanitization: The skill implements a robust 'Verification' phase (Step 3) where findings are adversarially checked by independent subagents against the source code, reducing the likelihood of successful manipulation by embedded comments.
Audit Metadata