The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/generate_schematic.py utilizes subprocess.run to invoke a local helper script, scripts/generate_schematic_ai.py, which manages the image generation workflow. This is a standard architectural pattern for modularizing the skill's functionality.
[EXTERNAL_DOWNLOADS]: The schematic generation tool in scripts/generate_schematic_ai.py performs network requests to the OpenRouter API (openrouter.ai). OpenRouter is a well-known service used to facilitate access to various large language models for generating and reviewing publication-quality diagrams.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection within the scripts/generate_schematic_ai.py script. (1) Ingestion Points: User-provided diagram descriptions are accepted as input. (2) Boundary Markers: No explicit delimiters or boundary instructions are utilized during string interpolation. (3) Capability Inventory: The tool performs network operations to an external API and handles local file writes for image data. (4) Sanitization: No input validation or escaping is applied to the user-supplied prompt before it is incorporated into the LLM system message. This is noted as an inherent risk in LLM-integrated features.

research-grants