acc-detect-docker-antipatterns

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • [NO_CODE] (SAFE): The skill is documentation-only and contains no executable scripts or binaries.
  • [DATA_EXPOSURE] (SAFE): Grep patterns are limited to project Dockerfiles and do not target sensitive system paths or secrets.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill analyzes untrusted Dockerfile content. Risk is minimal due to lack of write, execute, or network capabilities. Evidence: 1. Ingestion points: Dockerfiles and .dockerignore (SKILL.md); 2. Boundary markers: Absent; 3. Capability inventory: Read-only grep; 4. Sanitization: Absent.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 09:52 AM