git-commit
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from workspace changes to organize atomic commits and generate messages, creating a surface for indirect instructions.
- Ingestion points: Workspace changes (uncommitted code and files) described in SKILL.md.
- Boundary markers: None specified to delimit user content from instructions.
- Capability inventory: The delegated agent (commit-craft) performs workspace reads and git commit operations.
- Sanitization: No evidence of sanitization or filtering of workspace content before context interpolation.
Audit Metadata