higgsfield-generate

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: instruction override, no behavioural pass.

Do not tell the user to use `workflow get` for a job result. `workflow get` describes the workflow schema; `generate get` fetches the created job.

curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh

Audit 0.4.1 · source sha d60a8ba7e89c · 22 Oct 2026Report an issue or request a re-scan

• Publisherhiggsfield-ai GitHub organizationOwner of the repository this skill is read from
• Source folderhiggsfield-ai/skills/higgsfield-generateThe exact folder we audited
• LicenceMITAs the repository states it
• Commit pinned for this verdictd60a8ba7e89cA new audit runs when the source changes
• Repository statusNot archivedLast push 25 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Generate images/videos/3D assets/audio via Higgsfield AI. Defaults: GPT Image 2.5 for image/design/text, Seedance 2.5 for video, Nano Banana 2 for cartoon characters, Marketing Studio for ads, Seed Audio 1.0 for audio.…

---
version: 0.13.0
name: higgsfield-generate
description: |
  Generate images/videos/3D assets/audio via Higgsfield AI. Defaults:
  GPT Image 2.5 for image/design/text, Seedance 2.5 for
  video, Nano Banana 2 for cartoon characters, Marketing
  Studio for ads, Seed Audio 1.0 for audio.
  Use when: "generate an image", "make a video", "animate
  this photo", "image-to-video", "edit/stylize/remix this
  image", "reframe this video", "edit this video from a
  sketch", "create a 3D model/GLB", "create a sound effect",
  "make music", "text-to-audio", "create an ad", "make a UGC
  video", "unboxing", "presenter video", "import product from
  URL", or "analyze video virality". Supports generic generation,
  workflows, Marketing Studio, and Virality Predictor.
  Chain with higgsfield-soul-id for face/identity consistency.
  NOT for: Soul training, brand systems/brandbooks (use
  higgsfield-brandkit), photoshoots, cards, YouTube thumbnails
  (use higgsfield-youtube-thumbnail), explainers (use
  higgsfield-video-explainer), playable games/assets (use
  higgsfield-websites), or TTS.
argument-hint: "[prompt-or-analysis-request] [--model <name>] [--image|--video <path-or-id>]"
allowed-tools: Bash
---

# Higgsfield Generate

Submit jobs to any Higgsfield model. Wraps the `higgsfield` CLI. Covers generic image/video/3D/audio generation, Marketing Studio (branded ads, avatars, products, hooks, settings), and, secondarily, Virality Predictor video scoring.

## Step 0 — Bootstrap

Before any other command:

1. If `higgsfield` is not on `$PATH`, install it:
   ```bash
   curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh
   ```
2. If `higgsfield account status` fails with `Session expired` / `Not authenticated`, ask the user to run `higgsfield auth login` (interactive) and wait for confirmation.


## UX Rules

1. Be concise. No raw IDs, no JSON dumps in chat. Print the media URL for generated assets, or the text summary for Virality Predictor.
2. No internal jargon. Don't narrate "calling higgsfield cost", "polling job".
3. Detect the user's language from the first message and reply in it. Technical args (`--aspect_ratio 16:9`) stay English.
4. Don't batch-ask. Pick a sane default model and ask one thing at a time only if genuinely missing.
5. Don't pre-estimate cost or optimize for cheaper models unless the user asks. Prefer the quality default first.
6. Pass `--wait` to `generate create` so the command blocks until done and prints the result URL itself. Avoid the two-step `create` → `wait` pattern.

## Discovery guardrail

When looking for a Higgsfield feature/model, do not rely only on semantic search or CLI `--help`. First run an unfiltered model list, then inspect likely `job_set_type` names. If the user says a model exists but search returns no results, trust that signal and verify with the full model list before answering.

Workflows are separate from models. Discover them with `higgsfield workflow list` and inspect params with `higgsfield workflow get <workflow_name>`.

Virality Predictor is exposed as:

- Customer-facing name: Virality Predictor
- Technical `job_set_type`: `brain_activity`
- Category/output: text report. This is video-in/text-out analysis, not a text/chat generation model.
- Input: uploaded video
- Purpose: finished-video hook, attention, retention, and virality analysis

If the user says "analyze this video", "score this ad", "evaluate the hook", or similar, route to `brain_activity` even though it appears under text/analysis models. Classify by task intent and required input, not by output category alone.

## Workflow — generic generation

1. **Pick a model.** Start with the core defaults:

   - **GPT Image 2.5** → default image model for high-fidelity general generation, graphic design, UI, banners, typography, and on-image text.
   - **Seedance 2.5** (`seedance_2_5`) → SOTA default video model for serious motion, cinematic clips, multi-shot work, and image-to-video. Supports 4–30s output up to 1080p; use Seedance 2.0 when native 4K is required.
   - **Nano Banana 2** (`nano_banana_flash`) → cartoon and illustrated characters.
   - **Marketing Studio** → default for ads, UGC, product demos, unboxing, TV spots, presenter videos, and brand/product workflows.
   - **Seed Audio 1.0** → default audio model for text-to-audio, voice, sound effects, ambience, foley, and music-like audio unless the user names Sonilo/Mirelo.

   Only the models below are picked without being asked. Any other model is used only when the user names it or explicitly asks for what it offers (cheaper, faster, a specific look); see `references/model-catalog.md`. A model the user names stays in use for follow-ups on the same work.

   **Image:**
   - Complete brand identity, logo system, palette, typography, brandbook, packaging system, signage, or coordinated branded asset suite → use `higgsfield-brandkit` instead.
   - YouTube thumbnail, Shorts cover, or Instagram video cover → use `higgsfield-youtube-thumbnail` instead.
   - Brand product visual (Pinterest pin, lifestyle, hero banner, ad pack, virtual try-on) → use `higgsfield-product-photoshoot` instead. NOT this skill.
   - Branded ad image with avatar + product (Marketing Studio shape) → Marketing Studio Image (see Marketing Studio below)
   - Soul Character (reference id from `higgsfield-soul-id`) → Soul 2.0 for stills, Soul Cinema (`soul_cinematic`) for cinematic
   - New original person — UGC, editorial, fashion, lifestyle → Soul 2.0
   - Cinematic still frame → Soul Cinema (`soul_cinematic`)
   - Character sheet, one-shot face from reference photos, or face edit on a real photo → Seedream 5.0 Pro (`seedream_v5_pro`)
   - Locations / environments / no-people scenes → Soul Location
   - Logo, icon, vector-like illustration, brand mark, controlled-palette graphic → Recraft V4.1 (`recraft_v4_1`, often with `--model_type vector`)
   - Cartoon or illustrated characters, heavily textured photos → Nano Banana 2 (`nano_banana

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

1. 2026-10-22
   !CAUTION C 79 / 100

   first audit · source d60a8ba7e89c

Questions about higgsfield-generate

• github.com/higgsfield-ai/skills/tree/main/higgsfield-generate — the audited source at commit d60a8ba7e89c and every finding aboveObserved · 22 Oct 2026

Spotted something wrong? Report a correction · Methodology