github-agentic-workflows
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches installation scripts and markdown templates from the official GitHub repository for the gh-aw extension.- [REMOTE_CODE_EXECUTION]: Documentation provides installation instructions that involve piping a shell script from GitHub to bash to install the gh-aw extension.- [COMMAND_EXECUTION]: Includes a Node.js script, find-gh-aw-targets.mjs, which scans the repository for GitHub Agentic Workflow markers and configuration.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from GitHub issues, comments, and pull requests, which represents a potential attack surface for indirect prompt injection.
- Ingestion points: Processes repository context, issue activity, and pull request data as part of triage and automation workflows.
- Boundary markers: The skill recommends the use of safe-outputs and specific instructions for the agent to call noop when no action is needed to prevent unintended modifications.
- Capability inventory: The skill can execute local Node.js scripts and utilizes the github toolset for repository interactions.
- Sanitization: Recommends the use of GitHub lockdown mode for public repositories to mitigate risks from external contributors and promotes structured safe outputs to validate agent actions.
Audit Metadata