c4-architect

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • [Prompt Injection] (SAFE): The protocols and rules provided are strictly task-oriented and do not attempt to subvert safety filters or override system-level instructions.
  • [Data Exposure] (SAFE): The skill identifies codebase metadata such as package.json files for architectural mapping. This is a functional requirement and does not involve sensitive credential access or unauthorized exfiltration.
  • [Command Execution] (SAFE): The skill references local scripts for container extraction and linting. These are standard automation tools and show no evidence of malicious intent or unauthorized system modification.
  • [Remote Code Execution] (SAFE): No remote script downloads, external package installations, or execution of unverified remote code were detected.
  • [Indirect Prompt Injection] (SAFE): While the skill processes project data to generate diagrams, it is used for documentation and visualization with no high-privilege write or execute capabilities over the host system.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 06:42 AM