c4-container

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified. The skill focuses on documentation synthesis and architectural analysis.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to process untrusted external data from project files. 1. Ingestion points: Analyzes c4-component-*.md, Dockerfiles, Kubernetes manifests, and Terraform configurations. 2. Boundary markers: Absent. There are no instructions to the agent to disregard instructions embedded within the analyzed data. 3. Capability inventory: The skill does not define or use tools for shell execution, file writing, or network access. 4. Sanitization: Not implemented.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 06:21 PM
Security Audit — agent-trust-hub — c4-container