The Agent Skills Directory

[COMMAND_EXECUTION]: The skill implements a modular script architecture where scripts/generate_schematic.py utilizes the Python subprocess module to execute scripts/generate_schematic_ai.py. This is used to separate high-level logic from the AI-driven schematic generation and does not pose a security risk, as the execution is limited to internal skill files and avoids shell injection by passing arguments as a list.\n- [DATA_EXFILTRATION]: To generate scientific diagrams, the skill makes authorized network requests to the OpenRouter API (openrouter.ai). This communication is restricted to the intended functionality of creating schematics and follows security best practices by managing API keys through environment variables or local .env files rather than hardcoding them.\n- [EXTERNAL_DOWNLOADS]: The skill relies on standard, well-known libraries from the official Python Package Index (PyPI) such as pandas, numpy, scipy, lifelines, and matplotlib. These are established tools in the scientific community for data analysis and visualization.\n- [SAFE]: The skill includes defensive features like the scripts/validate_cds_document.py tool, which proactively checks clinical documents for quality, completeness, and HIPAA-compliant de-identification of sensitive patient information before finalization.

clinical-decision-support