github-actions
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a 'Secure by default' philosophy, explicitly instructing the agent to pin actions to full commit SHAs and utilize minimal 'permissions:' blocks in all generated workflows.
- [COMMAND_EXECUTION]: Utilizes standard, low-risk shell commands (
ls,cat,yq) for project stack detection and YAML validation. These operations are scoped to manifest files and are necessary for the skill's primary function. - [PROMPT_INJECTION]: No evidence of prompt injection or behavior override patterns. The skill includes robust internal checks to ensure generated workflows do not contain hardcoded secrets or unsafe script interpolations.
- [DATA_EXFILTRATION]: No unauthorized network operations or exfiltration patterns detected. Network activities are limited to standard GitHub CLI (
gh) operations for workflow management and debugging. - [CREDENTIALS_UNSAFE]: The skill correctly manages secrets by providing templates that use environment variables or OIDC, and includes verification steps to ensure no hardcoded credentials like API keys or tokens are present in generated output.
Audit Metadata