The catalogue/Higgsfield
Higgsfield
app.higgsfield
AI image and video generation with text-to-image, image-to-video, and Soul characters
Version2.0.0
Categoryai
AuthAPI key
Tools8
Transportstdio
LicenceMIT
01
What it does
The app, in its own words.
Generate AI images and videos with Higgsfield - text-to-image, image-to-video, and consistent character generation with Soul.
02
Tools
8 tools the agent can call.
Every one is declared with typed parameters and returns structured results, so the host can render them instead of narrating them.
- higgsfield-cancel-predictionCancel a queued or processing prediction. IRREVERSIBLE. Params: prediction_id (required).
- higgsfield-get-predictionGet the status and result of a prediction by ID. Params: prediction_id (required), wait? (poll for completion if still processing, default false).
- higgsfield-image-to-videoConvert an image to video with motion. Supports various motion styles and camera movements. Params: image_url (required), prompt (required), duration? (seconds, default 5), motion_strength? (0.0-1.0, default 0.5), camera_motion? ("none"|"zoom_in"|"zoom_out"|"pan_left"|"pan_right"|"tilt_up"|"tilt_down"), wait? (default false).
- higgsfield-list-modelsList available models and their capabilities. Params: type? ("all"|"text-to-image"|"image-to-video"|"text-to-video"|"image-to-image", default "all").
- higgsfield-runRun any model on Higgsfield. Supports text-to-image, image-to-video, and other AI generation models. Params: model (required — endpoint identifier, e.g. "bytedance/seedream/v4/text-to-image"), input (required — model parameters), wait? (poll for completion, default false).
- higgsfield-soul-generateGenerate images with consistent character appearance using Soul. Great for creating multiple images of the same character. Params: prompt (required), reference_image (required — URL of reference image), style?, resolution? ("1K"|"2K"|"4K", default "2K"), wait? (default false).
- higgsfield-text-to-imageGenerate images from text prompts using Seedream v4. Params: prompt (required), resolution? ("1K"|"2K"|"4K", default "2K"), aspect_ratio? ("1:1"|"16:9"|"9:16"|"4:3"|"3:4"|"21:9"|"9:21", default "16:9"), style?, wait? (default false).
- higgsfield-text-to-videoGenerate video from text prompt. Params: prompt (required), duration? (seconds, default 5), resolution? ("720p"|"1080p", default "1080p"), aspect_ratio? ("16:9"|"9:16"|"1:1", default "16:9"), wait? (default false).
03
Auth and permissions
What it asks for.
Declared in the manifest, in advance. The app cannot reach anything it did not ask for here.
User secrets
API_KEYAPI_SECRET
Reading further
The manifest and the source for Higgsfield live in the monorepo, alongside the other apps.