pr-pilot
Pass
Audited by Gen Agent Trust Hub on Jul 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified. The skill facilitates standard development operations using bundled local scripts and authenticated tools.
- [COMMAND_EXECUTION]: The skill uses bundled bash scripts (
ghchk,pilot.sh) to perform Git and GitHub CLI operations. These scripts are invoked with arguments derived from the repository state and user intent, and they do not download or execute remote code. - [DATA_EXFILTRATION]: Network operations are restricted to pushing changes to the repository's origin and creating PRs via the
ghtool. This interacts with GitHub, a well-known and trusted service, in a manner consistent with the skill's stated purpose. - [PROMPT_INJECTION]: The skill processes repository diffs to generate commit messages and PR descriptions. While this involves analyzing untrusted content from the codebase, the skill does not execute any instructions contained within that data, and no patterns intended to bypass AI safety filters were found.
Audit Metadata