docker-ci-cd

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official GitHub Actions and tools from trusted and well-known organizations, including 'docker', 'aquasecurity', and 'actions' (GitHub).
  • [CREDENTIALS_UNSAFE]: The skill appropriately instructs users to store credentials as CI secrets (e.g., GITHUB_TOKEN) and explicitly warns against passing sensitive tokens via build arguments to prevent leakage in image history.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted user-defined configuration inputs to generate CI workflows.
  • Ingestion points: User-provided inputs such as CI platform, registry targets, and architecture strings in SKILL.md.
  • Boundary markers: None present.
  • Capability inventory: The skill uses MCP tools (docker_listImages, docker_inspectImage) to interact with the local Docker daemon.
  • Sanitization: No explicit sanitization of user-provided configuration strings is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 07:30 AM
Security Audit — agent-trust-hub — docker-ci-cd