What problem does it solve? Vague requests like "make me an image" or "create a video" leave too many decisions open, leading to wrong formats, wrong models, and disorganized output files. This Skill turns any open-ended Higgsfield request into a finished, correctly named file by settling the four key decisions first: medium, model, format, and resolution. ## Core Features & Use Cases - Structured Intake: Asks four questions in a fixed order (image or video, which model, what format, what resolution), skipping anything the user already specified. - End-to-End Generation Workflow: Submits jobs through the Higgsfield MCP connector, polls for completion, downloads the full-resolution result, verifies file integrity, and renames files with a consistent slug-model-format-resolution pattern. - Batch Support: Runs multiple prompts in parallel groups and applies a custom base name across the whole set. - Use Case: A marketer says "generate a product shot for our launch." The Skill confirms medium, model, aspect ratio, and resolution, generates the image with Soul 2.0, and saves it as acme-launch-soul-2.0-1x1-4k-01.png in the chosen folder. ## Quick Start Ask the assistant to generate an image or video with Higgsfield, for example by saying "use Higgsfield to make a 9:16 product shot of our new earbuds in 4K."
