The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a local Python script (scripts/image_gen.py) that handles argument parsing, prompt augmentation, and API orchestration. The script manages local file operations for batch jobs and final image storage within designated project directories.
[EXTERNAL_DOWNLOADS]: The skill depends on standard, widely-used Python packages (openai, pillow) fetched from official package registries.
[DATA_EXFILTRATION]: The skill performs outbound network requests to OpenAI's official API endpoints. It explicitly implements safe credential handling by requiring the OPENAI_API_KEY to be provided via environment variables, with documentation warning the user against exposing the key in chat contexts.
[SAFE]: Analysis of the attack surface for indirect prompt injection: 1. Ingestion points: User-provided prompts enter the workflow through CLI arguments or JSONL input files processed by scripts/image_gen.py. 2. Boundary markers: The skill utilizes structured prompt templates (e.g., 'Primary request:', 'Constraints:') to isolate and clarify the role of user inputs. 3. Capability inventory: The skill is limited to writing image data to the local filesystem and performing network calls to the OpenAI API. 4. Sanitization: The script performs filename slugification and utilizes the official OpenAI SDK for all API communication.

imagegen