The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/generate_schematic.py executes a secondary internal Python script generate_schematic_ai.py using the subprocess.run method. This is used for delegating image generation tasks and follows best practices by using a list-based command structure to prevent shell injection.
[EXTERNAL_DOWNLOADS]: The skill interacts with external APIs at api.parallel.ai and openrouter.ai to perform research lookups and generate scientific schematics. These are documented external services essential to the skill's operation.
[INDIRECT_PROMPT_INJECTION]: The schematic generation tool interpolates user-provided descriptions directly into the prompt used by the AI reviewer model in scripts/generate_schematic_ai.py. While this creates an indirect injection surface, the impact is limited as the reviewer's output is used only for quality scoring and critique within a controlled workflow.
Ingestion points: The prompt argument in scripts/generate_schematic.py and scripts/generate_schematic_ai.py.
Boundary markers: None present in the interpolation of user input into the reviewer prompt.
Capability inventory: File system write access (for saving images), network access (for API calls), and local script execution via subprocess.
Sanitization: No specific sanitization or filtering is applied to the user-provided prompt before interpolation.

research-lookup