pr-description
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from git diffs and commit logs to generate summaries.\n
- Ingestion points: git diff and git log output gathered in workflow step 1 of repos/metamask-mobile.md.\n
- Boundary markers: None defined for separating untrusted git output from instructions.\n
- Capability inventory: The skill is authorized to write files to the local .agent/ directory.\n
- Sanitization: Branch names are sanitized by replacing slashes with dashes to mitigate path traversal risks.\n- [SAFE]: The skill uses standard development tools and commands (git, local file reads/writes) appropriate for its stated purpose of managing pull request metadata.
Audit Metadata