pr-finisher
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a constrained and safe operational environment for a GitHub Copilot cloud agent. It implements a 'no-merge' policy and explicitly prohibits the agent from triggering CI workflows, preventing unauthorized deployments.
- [COMMAND_EXECUTION]: The skill utilizes standard development tools, including the
ghCLI and project-specificmaketargets (e.g.,make fmt,make lint,make test). These commands are executed within the local context of the pull request and are necessary for the skill's function of validating code changes. - [DATA_EXFILTRATION]: The skill interacts with the GitHub GraphQL API and CLI to retrieve PR state. It caches this state locally in a temporary directory (
/tmp/gh-aw/pr-finisher/) for efficient processing within a single execution pass, which is a standard development practice and does not pose an exfiltration risk. - [PROMPT_INJECTION]: The skill includes comprehensive 'Hard rules' and 'Anti-patterns' that provide clear instructions to the agent to disregard potential manipulation attempts from external data, such as failing CI steps or malicious PR comments.
Audit Metadata