setup-pre-commit
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the repository's native package manager to install development dependencies such as Husky and lint-staged. It also involves executing validation checks against staged files to ensure the new hooks operate as intended.- [INDIRECT_PROMPT_INJECTION]: The skill functions by inspecting local repository files, including package.json, linter configurations, and existing hook scripts. This constitutes an ingestion surface where a repository containing malicious instructions in configuration files could attempt to influence the agent's behavior.
- Ingestion points: Local repository configuration files (package.json, .eslintrc, etc.) and existing git hooks (SKILL.md, Steps 1 and 2).
- Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions for the data being read from the project files.
- Capability inventory: The agent has the capability to modify package.json, install new packages, and create or overwrite script files.
- Sanitization: While no specific sanitization logic is provided, the instructions include safety-conscious directives such as 'preserve existing prepare scripts and hooks' and 'report absent checks rather than inventing scripts'.
Audit Metadata