The Agent Skills Directory

Data Exposure & Exfiltration (LOW): The script scripts/generate_image.py sends user-provided text prompts and base64-encoded image data to https://openrouter.ai/api/v1/chat/completions. This is the primary function of the skill, but users should be aware that data is transmitted to a third-party service.
Unverifiable Dependencies (LOW): The script depends on the requests library. While requests is a standard and widely-trusted Python package, it is an external dependency not included in the standard library. The script provides instructions for manual installation via pip.
Indirect Prompt Injection (LOW): The skill acts as a pass-through for user prompts and image data to an external AI model, creating a surface for indirect prompt injection.
Ingestion points: prompt argument and --input file path in scripts/generate_image.py.
Boundary markers: Absent. The text prompt is interpolated directly into the JSON payload for the API request.
Capability inventory: Performs network POST requests to OpenRouter and writes generated image files to the local file system via save_base64_image.
Sanitization: No sanitization or structural validation is performed on the input prompt or image data before transmission.

generate-image