treatment-plans

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFE
Full Analysis
  • [DYNAMIC_EXECUTION]: The generate_schematic.py script uses subprocess.run to invoke a secondary helper script for AI image generation. This is a functional requirement of the skill and uses safe practices by passing arguments as a list and avoiding shell=True.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with openrouter.ai to facilitate the generation and review of scientific schematics. This communication is restricted to sending prompts and receiving image data, utilizing environment variables for API key management rather than hardcoded credentials.
  • [INDIRECT_PROMPT_INJECTION]: The diagram generation feature accepts natural language input from the user which is then passed to an LLM. While this is an ingestion point for untrusted data, the risk is low as the impact is limited to the content of the generated image, and the process includes a quality-review iteration step.
  • [PRIVILEGE_ESCALATION]: The documentation includes instructions for users to manually install LaTeX dependencies using sudo tlmgr. These are standard setup procedures for the required LaTeX environment and do not indicate any attempt by the skill's scripts to automatically escalate privileges.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 02:14 AM