run-checks
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill performs several local shell operations using
git,pnpm, andemberto build, lint, and test the project. These commands are essential for the skill's primary purpose of ensuring code quality before a pull request. - Indirect Prompt Injection Surface: The procedure involves reading data from the repository's git history (commit SHAs and file paths) and using that information to scope further commands. While this is a standard automation pattern, it presents a surface where the agent processes untrusted metadata from the project's history.
- File System Modification: The skill is configured to run
pnpm lint:fixand formatting commands which modify files in the working directory. The instructions explicitly restrict the agent from making other modifications without user consent, which serves as a helpful safety boundary.
Audit Metadata