create-github-action-workflow-specification
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill acts as a documentation template for analyzing CI/CD workflows. It does not contain any instructions to execute shell commands, perform network requests, or access restricted system resources.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied GitHub Actions workflow files as input, creating a theoretical attack surface for indirect prompt injection. However, the risk is minimal and assessed as safe because the skill has no functional capabilities to execute code or exfiltrate data.
- Ingestion points: Untrusted content is ingested from the
${input:WorkflowFile}variable specified in SKILL.md. - Boundary markers: The instructions do not define clear delimiters to separate the ingested workflow content from the agent's instructions.
- Capability inventory: The skill is restricted to text and diagram (Mermaid) generation; no tools, subprocesses, or network APIs are utilized.
- Sanitization: There are no explicit instructions for the agent to sanitize or ignore malicious instructions embedded within the workflow files.
Audit Metadata