Higgsfield Seedance

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: no behavioural pass.

Audit 0.4.1 · source sha 9e3a2f685c72 · 31 Oct 2026Report an issue or request a re-scan

• PublisherOSideMedia GitHub userOwner of the repository this skill is read from
• Source folderOSideMedia/higgsfield-ai-prompt-skill/skills/higgsfield-seedanceThe exact folder we audited
• LicenceMITAs the repository states it
• Commit pinned for this verdict9e3a2f685c72A new audit runs when the source changes
• Repository statusNot archivedLast push 6 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Rewrites scene descriptions using professional cinematography language, structures prompts with a six-slot formula (camera + subject + action + setting + style + lighting), and diagnoses content filter rejections via a…

---
name: higgsfield-seedance
description: "Rewrites scene descriptions using professional cinematography language, structures prompts with a six-slot formula (camera + subject + action + setting + style + lighting), and diagnoses content filter rejections via a preflight linter. Use whenever the user asks for a Seedance 2.0 / Seedance Pro prompt, describes a scene for Seedance generation, mentions Seedance, reports a Seedance generation failure or flagged prompt, or is burning credits on Seedance regenerations."
user-invocable: true
metadata:
  tags: [higgsfield, seedance, seedance-2.0, seedance-pro, content-filter, prompt, director, flagged]
  version: 1.16.0
  updated: 2026-09-26
  parent: higgsfield
---

# Higgsfield Seedance Director

## QUICK FACTS
*Generated-checked block (scripts/build_index.py verifies anchors). Read the linked sections for full context — these lines are routing aids, not the rules themselves.*
- The filter is an LLM reading full-scene intent, not a keyword blacklist — describe a SCENE, not a subject; fix the voice first [→](#the-filter-model--read-this-first)
- Instant fail (<10s) = filter rejection; delayed fail (>30s) = infra/complexity — never regenerate an instant fail unchanged [→](#instant-fail-vs-delayed-fail--the-diagnostic)
- Six slots, in order: Camera + Subject + Action + Setting + Style + Lighting; missing 3+ slots is where flags come from [→](#the-seedance-prompt-formula)
- Empirical prompt-craft laws: 50–80-word attention sweet spot (front-load the load-bearing element), name a director/lens not "cinematic", "fast" degrades motion, no negative prompts in the body, unidirectional motion chains + named camera endpoint + detail scale follows shot size [→](#prompt-craft-laws)
- Five prompt modes: Reference-Based / Continuation / Expand Shot / Edit Shot / Transformation — pick the mode before writing [→](#seedance-20-prompt-modes)
- [OFFICIAL] block scaffold for production prompts: SCENE CONTEXT → ... → POSITIVE LOCKS, distributed style on standalone briefs (connected shotlists glue the compiled Style Prefix verbatim instead), FOV in degrees only, CAMERA block 3rd, cut ladder oner / CUT n / timed / freestyle [→](#official-prompt-architecture--the-block-scaffold)
- [FIELD] 13-project corpus calibration: word length scales with register (218w → 2,059w medians — the 50–80w sweet spot is single-shot-only), video briefs hand-authored (`enhance_prompt` off), Style Prefix = per-project constant compiled into home blocks [→](#field-calibration--the-13-project-production-corpus)
- [FIELD] Three "helpful-instinct" drift sources, each with a standing lock: environment invention (#1, above character drift), character-height equalization, scale drift on wides [→](#positive-locks)
- Build-safe construction for crowds/destruction/creatures: evacuated cities, contained fights ("stays at the sea surface"), the safe benchmark scene [→](#build-safe-construction--crowds-destruction-creatures)
- Extend an existing clip: attach it as a video reference + open with "The scene continues." — match source resolution AND duration; chain cap ~2 (hard 3), then re-anchor from ORIGINAL references [→](#extension-prompting--video-reference-continuation)
- **This file is Seedance 2.0.** For **2.5** — four modes incl. `video_edit` / `video_extension`, 4–30s, 30/10/10 references, first-last frames in `omni_reference`, up to 1080p (no 4K, no genre) — use `../higgsfield-seedance-2-5/SKILL.md`
- Tutorial-demonstrated patterns (reference-role vocabulary incl. VARIETY reference, SCREEN REALISM + duration-match composites, 60:30:10 grade, red-arrow prop annotation): `PRODUCTION-PATTERNS.md` in this directory
- [OFFICIAL] Feature-film pipeline (asset construction, per-scene GEO SPATIAL LAYOUT, the position-fixing first second, dialogue construction, ban dictionary, the 10–15 iteration rule, crowds / giants / thresholds): `HELL-GRIND.md` in this directory
- Performance — objective, obstacle, tactics, beats, subtext, eye life, the acting master profile: `../higgsfield-acting/SKILL.md`
- Hard engine rules (age-blind, exit-frame = cut, off-screen = nonexistent, no reflections, ≤3 tracked characters, double-contrast cuts) + high-risk shot table: `ENGINE-RULES.md` in this directory
- **Render failures** (the take came back wrong, not flagged) — 15 named failures with symptom · mechanism · counter, incl. action-reversal fill, filler-babble, truncated action, mimed manipulation, orphan limbs / the solo third hand, walking + cross-shot camera speed, fights as separate clips — plus a self-repair checklist: `FAILURE-MODES.md` in this directory
- [FIELD] **Bake it into the asset**: a property that drifts however well it is written (a lens character, a grain, a colour cast) moves one step earlier, into the plate — then, in one studio's practice, leaves the video prompt entirely (OPEN: Hell Grind keeps the look in the prompt too, P2-6); bake only what should stay constant [→](#bake-it-into-the-asset-when-the-prompt-will-not-hold-it)
- [FIELD] **Depth map** as a reference role — greyscale, light = near; a geometry input that stops a space re-planning itself between shots, carries no style [→](#depth-map)
- Contested doctrine (character-text volume with a reference attached, anamorphic words vs a baked plate, `NO BGM` vs `No music.`, the stop numbers) is recorded — not silently resolved — in `../shared/house-rulings.md`; label meanings in `../shared/provenance.md`
- Reference roles: Character / Last-Frame / Environment / Prop — role determines what the prompt may re-describe [→](#reference-roles)
- Working modes: Exploration / Continuation / Bridging / Repair (distinct from prompt modes) [→](#working-modes-vs-prompt-modes--two-taxonomies)
- Layer 1 briefing vs Layer 2 production prompt — never paste Layer 1 into the prompt box [→](#two-layer-prompt-authoring)
- Native **4K** is available in `mode=std` only; `mode=fast` (Seedance 2.0 Fast) caps at 480p/720p — in Cinema Studio the model is still capped at 1080p [

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.

git clone https://github.com/OSideMedia/higgsfield-ai-prompt-skill
git -C higgsfield-ai-prompt-skill checkout 9e3a2f685c72

Works in

1. 2026-10-31
   ✓SAFE B 89 / 100

   first audit · source 9e3a2f685c72

Questions about Higgsfield Seedance

• github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skill — the audited source at commit 9e3a2f685c72 and every finding aboveObserved · 31 Oct 2026

Spotted something wrong? Report a correction · Methodology