ci-cd-and-automation
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or security vulnerabilities were detected in the skill's instructions or configuration templates.
- [CREDENTIALS_UNSAFE]: The skill adheres to security best practices by explicitly advising against hardcoding credentials. It provides examples using GitHub Secrets (e.g.,
${{ secrets.CI_DB_PASSWORD }}) and environment variable management for production and CI environments. - [EXTERNAL_DOWNLOADS]: The configuration examples utilize well-known, industry-standard tools and GitHub Actions, such as
actions/checkout,actions/setup-node,playwright, andvercel. These references are documented for legitimate build and deployment purposes. - [INDIRECT_PROMPT_INJECTION]: The skill describes a feedback loop where CI failures are provided to the agent for remediation. While this involves processing external log data, the skill provides a clear instructional template to guide the agent, which is a standard and safe development workflow pattern.
Audit Metadata