modelslab-image-editing

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt's examples pass API keys as function arguments and embed them verbatim into request JSON (and usage shows literal "your_api_key"), which requires the LLM to output or include secret values directly in generated requests/code, creating exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The skill accepts and posts arbitrary image URLs (e.g., the image_url parameter in functions like remove_background, edit_image_with_ai, outpainting, and other endpoints) to ModelsLab's public API, meaning it fetches and processes untrusted third‑party content from the open web as part of its workflow.