comfyui-import

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

• Third-party content exposure detected (high risk: 0.70). The skill ingests a user-supplied ComfyUI workflow JSON (including prompts and other text fields) and then writes/derives pipeline specs from that JSON; since the workflow content is not guaranteed to be authored by the operating user, it can include outsider-authored free text (e.g., prompts) that becomes LLM-readable context during parsing/processing.