model-office

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The orchestration script scripts/run.py initiates the local FastAPI backend (scripts/server.py) using subprocess.Popen. This is a functional requirement to host the pixel-art visualization dashboard on http://127.0.0.1:8787.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined tasks which are interpolated into a dispatcher prompt. While the skill lacks explicit boundary markers or sanitization for this untrusted data, the capabilities are restricted to generating visualization events and making standard API calls to configured LLM providers (Anthropic, OpenAI, and Perplexity).
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 03:14 PM
Security Audit — agent-trust-hub — model-office