together-images

Pass

Audited by Gen Agent Trust Hub on Apr 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates legitimate image generation tasks using the official Together AI SDK and well-known platform resources.
  • [EXTERNAL_DOWNLOADS]: Fetches image assets and LoRA adapters from well-known and trusted domains including Together AI, Hugging Face, Pixabay, and Pexels.
  • [COMMAND_EXECUTION]: Contains scripts that write generated image data to the local file system using standard file operations. This behavior is essential for the skill's functionality and is restricted to image output.
  • [PROMPT_INJECTION]: The skill defines a processing surface for user-supplied prompts and image URLs (Category 8: Indirect Prompt Injection). 1. Ingestion points: User prompts and image URLs processed in scripts/generate_image.py and scripts/kontext_editing.py. 2. Boundary markers: Absent; instructions do not explicitly frame external inputs with protective delimiters. 3. Capability inventory: Local file writing and network requests via standard libraries. 4. Sanitization: Scripts do not implement explicit input sanitization for file paths, relying on standard library behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 3, 2026, 02:11 AM