auditing-gha-settings
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes the
gh(GitHub CLI) tool to retrieve repository-level Actions permissions and allowlists via the GitHub API. This operation is limited to reading metadata for comparison against a local baseline. - [DATA_EXFILTRATION]: The skill retrieves sensitive repository configuration data (GitHub Actions allowlists and permissions). Analysis of
run.mtsconfirms that this data is processed locally and logged to the console, with no network operations sending the data to external third-party domains. - [PROMPT_INJECTION]: The skill ingests untrusted data from GitHub API responses. This creates an indirect prompt injection surface; however, the skill logic is restricted to comparison and logging, and does not dynamically execute instructions or interpolate data into command execution in a way that could be exploited.
Audit Metadata