optimise-github-actions
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The
scripts/measure.mjsscript usesnode:child_process.execFileto invoke the GitHub CLI (gh) for retrieving repository metadata, workflow run histories, and job details.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests and displays untrusted data from the GitHub API, creating a surface for potential indirect prompt injection attacks.\n - Ingestion points: Workflow names, job names, branch names, and status labels are fetched via the GitHub API and stored in
jobs.jsonor printed to the console.\n - Boundary markers: Not present. Data is processed and displayed without specific delimiters or instructions to ignore embedded commands.\n
- Capability inventory: The skill possesses the capability to execute commands via the
ghtool and to modify local files, including workflow configurations in the.github/workflows/directory.\n - Sanitization: The script does not perform sanitization or filtering of the strings returned by the GitHub API before presenting them to the user or agent.\n- [EXTERNAL_DOWNLOADS]: The instructions suggest downloading the
actionlintutility to validate workflow changes if the tool is not already installed in the environment.
Audit Metadata