github-actions-templates
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides secure GitHub Actions workflow patterns. It correctly demonstrates security best practices by pinning third-party actions to specific commit SHAs (e.g., actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd) to prevent supply chain attacks. It also restricts job permissions to 'contents: read' and disables credential persistence in the checkout step. The instructional content in references/common-workflows.md and resources/implementation-playbook.md consistently reinforces the importance of using unprivileged jobs for untrusted code and separating publication paths.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest data from the target repository.
- Ingestion points: SKILL.md (instructions to inspect repository scripts and runtime configurations).
- Boundary markers: Absent in the provided templates.
- Capability inventory: Writing workflow files to the repository.
- Sanitization: Absent; the skill relies on the agent's manual review of the repository content.
- Assessment: This surface is inherent to the skill's purpose as a workflow generator and does not pose a significant risk given the instructional focus on security reviews.
Audit Metadata