committing-with-guidelines
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various git commands and arbitrary repository-specific tests or linters to validate commits. Evidence includes the direct invocation of
git status,git diff,git add,git commit,git log,git rebase, andgit push, as well as instructions to 'Run the repository's relevant tests, linters, or checks.' - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the git repository and uses it to perform actions, creating a surface for potential injection. 1. Ingestion points: The agent reads repository diffs and worktree status in SKILL.md. 2. Boundary markers: The instructions do not include markers or warnings to disregard instructions found within the repository code. 3. Capability inventory: The agent has shell command execution capabilities for git operations and project scripts. 4. Sanitization: No sanitization of the repository content is specified before it is processed by the agent.
Audit Metadata