git-pr-workflows-git-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze uncommitted code changes from the local repository, which constitutes untrusted external data. This data could contain instructions designed to influence the behavior of the agents processing it.
- Ingestion points: Uncommitted changes and file lists are processed in Phase 1 (Code Quality Assessment) and Phase 3 (Change Analysis and Categorization).
- Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the ingested code.
- Capability inventory: The workflow uses the analyzed data to generate branch management commands, commit messages, and PR descriptions (Phases 3, 4, and 5).
- Sanitization: No explicit sanitization or validation of the code input is mentioned in the instructions.
Audit Metadata