Video Generation
bytedance/deer-flow
Generates short videos from a structured JSON prompt, optionally guided by a reference image used as the first or last frame.
A skill your agent uses when the user asks about visual styles, aesthetics, color grades, film looks, or how to set the tone and atmosphere of a Higgsfield generation.
$ npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install OSideMedia/higgsfield-ai-prompt-skill higgsfield-style --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/OSideMedia/higgsfield-ai-prompt-skill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/higgsfield-style .claude/skills/higgsfield-style && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "higgsfield-style" agent skill from https://github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skills/higgsfield-style into .claude/skills/higgsfield-style/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "higgsfield-style", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skills/higgsfield-styleType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install OSideMedia/higgsfield-ai-prompt-skill higgsfield-style --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/OSideMedia/higgsfield-ai-prompt-skill.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/higgsfield-style .agents/skills/higgsfield-style && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "higgsfield-style" agent skill from https://github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skills/higgsfield-style into .agents/skills/higgsfield-style/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "higgsfield-style", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install OSideMedia/higgsfield-ai-prompt-skill higgsfield-style --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/OSideMedia/higgsfield-ai-prompt-skill.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/higgsfield-style .cursor/skills/higgsfield-style && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "higgsfield-style" agent skill from https://github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skills/higgsfield-style into .cursor/skills/higgsfield-style/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "higgsfield-style", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/OSideMedia/higgsfield-ai-prompt-skill.git --path skills/higgsfield-style--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install OSideMedia/higgsfield-ai-prompt-skill higgsfield-style --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/OSideMedia/higgsfield-ai-prompt-skill.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/higgsfield-style .gemini/skills/higgsfield-style && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "higgsfield-style" agent skill from https://github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skills/higgsfield-style into .gemini/skills/higgsfield-style/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "higgsfield-style", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install OSideMedia/higgsfield-ai-prompt-skill higgsfield-styleInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/OSideMedia/higgsfield-ai-prompt-skill.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/higgsfield-style .github/skills/higgsfield-style && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "higgsfield-style" agent skill from https://github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skills/higgsfield-style into .github/skills/higgsfield-style/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "higgsfield-style", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install OSideMedia/higgsfield-ai-prompt-skill higgsfield-style --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/OSideMedia/higgsfield-ai-prompt-skill.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/higgsfield-style .opencode/skills/higgsfield-style && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "higgsfield-style" agent skill from https://github.com/OSideMedia/higgsfield-ai-prompt-skill/tree/main/skills/higgsfield-style into .opencode/skills/higgsfield-style/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "higgsfield-style", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
higgsfield-styleA skill your agent uses when the user asks about visual styles, aesthetics, color grades, film looks, or how to set the tone and atmosphere of a Higgsfield generation.
Higgsfield Style is an agent skill from OSideMedia/higgsfield-ai-prompt-skill. Use when the user asks about visual styles, aesthetics, color grades, film looks, or how to set the tone and atmosphere of a Higgsfield generation.
Its SKILL.md is about 3.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Media & Creative, covering AI video generation. The repository describes itself as: Claude AI skill for cinematic Higgsfield AI prompts — 32 sub-skills covering Seedance 2.5 (omni-reference, video edit + extend) and 2.0, the Hell Grind feature-film pipeline, an… The licence is MIT.
Read from SKILL.md and the folder at commit 7075497. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Higgsfield Style loads about 3.9k tokens when it runs. Until then it costs about 41 tokens; SKILL.md has 1,781 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from OSideMedia/higgsfield-ai-prompt-skill at commit 7075497, republished under its MIT licence (© OSideMedia). 1,781 words, ~3,872 tokens.
.claude/skills/higgsfield-style/SKILL.md (or your agent's skills folder).These five styles are Higgsfield's named presets. Reference them by exact name.
Look: Polished, high-contrast, vivid colors, balanced exposure — modern blockbuster Best for: Drama, action, narrative films, commercials, any professional content Color tendency: Rich, saturated, clean Prompt phrase: "Style: Cinematic" Pair with: Kling 2.6/3.0, Dolly In, Arc, Crane Up
Example: A detective walks through a night market.
Style: Cinematic. Cold blue shadows, warm amber market stall light.
Shallow depth of field. 16:9.Look: Retro videotape grain, color bleed, slight scanlines, analog imperfection Best for: 80s/90s nostalgia, horror, thriller, retro music videos, flashbacks Color tendency: Slightly washed out, warm yellows and reds, low contrast Prompt phrase: "Style: VHS" Pair with: Handheld camera, Wan 2.6 or Wan 2.5 (2.5 is not in the API catalog, 2026-09-26 — verify in the live UI), any horror preset
Example: Friends at a house party in 1987.
Style: VHS. Warm, grainy, slightly overexposed. 4:3 ratio.Look: Warm film grain, soft vignette, muted colors, home-movie feel Best for: Personal stories, romance, nostalgia, indie films, family moments Color tendency: Warm, golden, slightly faded Prompt phrase: "Style: Super 8MM" Pair with: Handheld, natural light descriptions, intimate scenes
Example: A couple dancing in a sunlit backyard in the 1970s.
Style: Super 8MM. Warm grain, soft vignette edges. 4:3.Look: Ultra-wide aspect ratio (2.35:1), horizontal lens flares, slight barrel distortion, epic scale — classic Hollywood widescreen Best for: Action, epic fantasy, war films, sweeping landscapes, high drama Color tendency: High contrast, deep blacks, rich highlights Prompt phrase: "Style: Anamorphic" or "Style: Anamorphic, 2.35:1 widescreen" Pair with: Crane Up, 360 Orbit, Super Dolly Out, Seedance 2.0 / Minimax Hailuo 2.3
Example: An army marches across a frozen plain at dawn.
Style: Anamorphic, 2.35:1. Deep blue-grey tones. Lens flare on the rising sun.Look: Non-representational, surreal color schemes, unconventional shapes, artistic Best for: Music videos, conceptual art, dream sequences, experimental content Color tendency: Vivid, unexpected, driven by concept not realism Prompt phrase: "Style: Abstract" Pair with: Wan 2.6 or Wan 2.5 (2.5 is not in the API catalog, 2026-09-26 — verify in the live UI), Portal, Multiverse, Glitch presets
Example: Fractured geometric shapes pulse to music in a void.
Style: Abstract. Electric blue and magenta on black. 1:1 ratio.Use these in any prompt regardless of style preset:
| Mood | Color grade description |
|---|---|
| Cold thriller | "Teal and orange, desaturated, high contrast" |
| Warm nostalgia | "Golden hour amber, soft shadows, low contrast" |
| Cyberpunk | "Neon magenta and cyan, deep shadows, HDR" |
| Horror | "Sickly green-yellow, crushed blacks, desaturated" |
| Romance | "Soft warm pink-gold, lifted shadows, dreamy" |
| Documentary | "Neutral, natural light, no grade" |
| Epic fantasy | "Rich jewel tones, deep shadows, volumetric light" |
| Noir | "High contrast black and white, or near-monochrome" |
| Sci-fi cold | "Ice blue and silver, stark white light" |
| Post-apocalyptic | "Desaturated orange and brown, dust haze" |
| Type | Description | Best for |
|---|---|---|
| Golden hour | Warm directional light just after sunrise or before sunset | Romantic, epic, beautiful |
| Overcast | Soft diffused light, no shadows | Documentary, emotional, grounded |
| Neon | Colored artificial light from signs/screens | Cyberpunk, night scenes, urban |
| Volumetric | Light rays visible through atmosphere (fog/dust) | Fantasy, cinematic, atmospheric |
| Practical only | All light comes from sources visible in frame (lamps, fire, screens) | Realism, noir, intimate |
| Side-lit | Single strong light from one side creating deep shadow | Drama, tension, portrait |
| Backlit | Subject silhouetted or rimlit from behind | Mystery, romance, epic reveal |
| Low key | Mostly dark with small pools of light | Horror, thriller, noir |
| High key | Bright, even, minimal shadows | Comedy, commercial, lifestyle |
Specific lighting setups that AI models respond to well. Use these terms directly in your prompts for precise control over how light shapes the scene.
| Technique | Effect | Best for |
|---|---|---|
| Rembrandt lighting | Triangle of light on the shadowed cheek, one eye lit | Portrait drama, character intros, moody interviews |
| Butterfly / Paramount lighting | Overhead light casting a shadow under the nose | Glamour, fashion, beauty shots, classic Hollywood |
| Split lighting | Half the face lit, half in complete shadow | Duality, inner conflict, villain reveals |
| Rim lighting / backlit | Edge glow outlining the subject's silhouette | Mystery, epic reveal, separation from background |
| Motivated lighting | Light source visible or implied in the scene (lamp, window, fire) | Realism, narrative grounding, naturalistic drama |
| Practical lighting | In-scene light sources (neon signs, candles, screens) | Night scenes, cyberpunk, intimate realism |
| Chiaroscuro | Extreme contrast between light and dark areas | Renaissance feel, high drama, painterly compositions |
| High-key | Bright, minimal shadows, even illumination | Comedy, commercial, lifestyle, clean aesthetic |
| Low-key | Deep shadows dominate, small pools of light | Noir, thriller, horror, psychological tension |
| Golden hour / Magic hour | Warm amber directional light, long soft shadows | Romance, beauty, epic landscapes, emotional beats |
| Blue hour | Cool steel-blue ambient light just after sunset | Melancholy, transition, quiet tension, urban solitude |
| Harsh midday sun | Hard overhead light, strong defined shadows | Desert, confrontation, exposed vulnerability |
| Overcast diffused / softbox | Even soft light, no hard shadows | Portraits, documentary, grounded realism |
| Ratio | Name | Best for |
|---|---|---|
| 16:9 | Widescreen | Standard video, YouTube, film |
| 9:16 | Vertical | TikTok, Instagram Reels, Shorts |
| 2.35:1 | Anamorphic | Epic cinema, maximum widescreen drama |
| 1:1 | Square | Instagram posts, artistic |
| 4:5 | Portrait | Instagram feed, social portrait |
| 4:3 | Classic TV | Retro, VHS feel, vintage |
Negative constraints: For texture/lighting artifacts (flickering textures, style ignored, over-lit output, color grade inconsistency) and their prevention phrases, see
../shared/negative-constraints.md— Texture/Lighting Artifacts section.
These principles apply to Cinema Studio 3.0's generation engine (Business/Team plan only) and complement the style vocabulary above.
ONE primary style anchor beats five adjectives. Beyond 2–3 style tokens, model attention dilutes and the output becomes generic.
Wrong: Style: cinematic, anamorphic, moody, atmospheric, dramatic lighting, film grain, desaturated, noir-inspired, high contrast, vintage feel
Right: Style: anamorphic, subtle grain, muted palette
Pick your anchor (the single most important style element), add 1–2 supporting tokens, stop.
Every generated video is "cinematic" by default. The word adds zero information. Replace it with a specific lens or contrast description:
shallow depth of field, warm highlights, cool shadowsanamorphic, 2.35:1, horizontal lens flares35mm film stock, natural grain, Kodak Portra paletteVisual style references beat descriptive text. One reference image/video carries more style information than 10 descriptor words:
Match the visual style, color grading, and film texture of @Video1.
A woman walks through autumn leaves in a park.
Camera: slow tracking alongside her.Use style references for: color grading, film stock emulation, lighting mood, texture quality, era-specific looks.
When prompting CGI or product renders, specify 2–4 material properties per surface to avoid the default "plastic sheen":
| Property | Options | Example |
|---|---|---|
| Base | metal, glass, fabric, ceramic, wood, leather | brushed stainless steel |
| Roughness | matte, satin, glossy, mirror | satin finish |
| Imperfection | scratches, dust, wear, fingerprints, patina | fine scratches from use |
| Edge | beveled, sharp, rounded, chamfered | soft rounded edges |
Example: A matte ceramic vase with hairline cracks and a subtle patina, soft rounded rim, resting on rough-hewn oak.
Don't just name the decade — specify the materials and lighting of the era:
Kodachrome warm tones, wood paneling, orange shag carpet, tungsten bulbs casting amber lighthigh-contrast black and white, Venetian blind shadows, fedora silhouettes, wet asphalt reflecting streetlampsHi8 camcorder grain, autofocus hunting, timestamp overlay, oversaturated greens[FIELD — 13-project community harvest, 2026-07-18] — "photoreal cinematic"
is not one register. Production projects sit at named poles of a dial, and
the pole decides the whole camera grammar:
| Register pole | Camera grammar | Texture stack |
|---|---|---|
| Film register (drama/romance features) | Handheld breathing, off-level allowed, natural motion | 35mm grain, milky low contrast, gate weave, halation, "corners as bright as center" |
| Broadcast-TV register (early-2000s drama) | Locked-off tripod, frontal compositions, held reaction beats, ONE camera move per shot maximum | Heavy soft diffusion, blooming highlights, telecine grain, flat neutral daytime |
| Stop-motion / hand-animated | Stepped motion — "true 12fps, animated on twos: each pose held two frames then snapping, never gliding" + constant painterly boil | Split-motion rule: atmosphere (snow, breath-vapor, smoke) moves SMOOTHLY while figures step on twos — an explicit constraint that fights the video model's default smoothness |
| Anime / cel | Per-shot named camera, fewer physics/skin blocks (the template contracts) | "Cel-shaded 2D, clean flat fills, two-to-three value cel shading — not painterly, not 3D, no CGI smoothness" |
The style-anchor slot swaps vocabulary by register. Same slot, different language: photoreal anchors on a DP/director look (short-form only — block prompts describe the look instead, per the seedance measurable-language rules) · anime anchors on an art era ("early 2000s retro anime, vintage cel proportions") · stop-motion anchors on medium physics ("12fps on twos, painterly boil"). Never carry one register's anchor into another.
One saturated accent color, reserved for the story. The corpus-wide color discipline: the plot-critical prop owns the only saturated accent in a muted grade (an acid-green device, a chartreuse remote LED) — and the same reservation works temporally ("golden hour is reserved for the twist"; everything before it stays flat neutral). State the reservation explicitly.
[OFFICIAL — Higgsfield cinematic-prompt-builder skill, 2026-07] — starting
recipes, each a distinct render contract. Combine and deviate freely; the
load-bearing part of each is the is / is-NOT declaration and the
shots-per-15s shape:
| Recipe | Look declaration | Structure + signature | Audio |
|---|---|---|---|
| Live-action epic | 8K photoreal, anamorphic, fine grain — "photoreal, NOT 3D/game" | Oner or multishot; one "hook" event triggers one held slow-mo beat, then snaps back; scale contrast (tiny figures vs vast subject) | Diegetic; slow-mo drops to muffled vacuum + heartbeat, snaps back with a whoosh |
| 3D animated feature | Vibrant glossy CGI, Pixar-quality stylized, subsurface fur/skin | 6 shots / 15s montage — "don't use one camera angle"; per-shot spoken lines with delivery described; land a visual gag | Light score allowed; often still SFX-forward |
| Game cutscene | UE5 real-time in-engine render — "hyperreal but unmistakably GAME-rendered, NOT film" | 3 shots / 15s, hard cuts; screen-pinned HUD in a locked accent hex, every element enumerated with exact text — HUD never parallaxes | No music; SFX + subtle UI ticks |
| Gameplay footage | AAA arcade (racing) — hyper-saturated engine render, "in-game, NOT a film plate" | Rock-steady chase-cam locked to vehicle, NO slow-mo anywhere, live HUD (RPM, boost, minimap) | Engine, gears, boost, tyre screech, UI beeps |
| FPV / POV oner | One unbroken take, photoreal macro, anamorphic | Beats by timestamp in one paragraph; one named speed ramp (240fps bullet-time beat → snap to 24fps); macro brushes against surfaces | Slow-mo swaps ambience for amplified rhythmic sound |
| Product / commercial | 8K editorial, deep black crush, per-section color worlds | 8–10 short sections, hard match-cuts on action; rapid 0.3s macro montage; resolves on a hero packshot (product in one third, dark negative space, slow dolly, particle drift, hold) | No music/VO/text — product-handling SFX only |
| VFX composite (on a source clip) | Photoreal practical-FX realism; composition + grade INHERITED from the source | INPUT LOCK: plate preserved unchanged, camera inherited exactly; only additions = new element + its light + contact shadows; parallax-locked to an anchor plane | New element's diegetic sounds over preserved room tone |
| Kaiju / creature | 8K photoreal anamorphic, ORIGINAL design ("not based on any franchise") | ~6 cuts, per-cut OPTICS; the detail that opens CUT 1 is the one CUT 2 zooms into; containment rule ("stays at the sea surface, never airborne") grounds physics and scale | Diegetic only — rumble, roars, water-burst, subsonic boom |
Full engine-side counterparts: the HUD recipe and INPUT LOCK live with
worked patterns in ../higgsfield-seedance/SKILL.md and
../higgsfield-seedance-vfx/SKILL.md; the containment rule doubles as the
build-safe construction law (../higgsfield-seedance/SKILL.md § Build-safe
construction).
higgsfield-camera — Camera controls to pair with styleshiggsfield-mixed-media — Artistic style overlays (non-photorealistic)higgsfield-moodboard — Moodboard + Soul Hex for project-level style lockinghiggsfield-cinema — Cinema Studio built-in color grading suitetemplates/ — Annotated genre-specific prompt templates with style examples© OSideMedia, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Just SKILL.md in skills/higgsfield-style of OSideMedia/higgsfield-ai-prompt-skill.
Open the folder on GitHubat commit 7075497
Higgsfield Style next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Higgsfield Style this skillOSideMedia/higgsfield-ai-prompt-skill | 713 | — | ~3.9k | Automated safety check: Pass | MIT | |
| Video Generationbytedance/deer-flow | 84k | 3 repos | ~1.4k | Automated safety check: Pass | MIT | |
| Video Cover Imageitwanger/toBeBetterJavaer | 18k | — | ~3.3k | Automated safety check: Pass | None | |
| Seedancesongguoxs/seedance-prompt-skill | 2.9k | 1 repos | ~2.5k | Automated safety check: Pass | None | |
| HyperFrames Video Entry Pointheygen-com/hyperframes | 60k | 3 repos | ~5.2k | Automated safety check: Pass | Apache-2.0 | |
| Lanshu Create AI Presenter Videocclank/lanshu-create-ai-presenter-video | 2.6k | — | ~3.6k | Automated safety check: Pass | MIT |
bytedance/deer-flow
Generates short videos from a structured JSON prompt, optionally guided by a reference image used as the first or last frame.
itwanger/toBeBetterJavaer
Generate matched 3:4, 16:9, and 4:3 short-video cover images from toBeBetterJavaer video scripts or AI/Java technical topics.
songguoxs/seedance-prompt-skill
This skill should be used when the user asks to "generate video prompts", "create Seedance prompts", "write video descriptions", mentions "Seedance", "seedance", "即梦", "即梦平台", "视频提示词", "视频生成"…
heygen-com/hyperframes
Entry point for making, editing and rendering videos from HTML compositions with HyperFrames, routing each request to the right workflow.
cclank/lanshu-create-ai-presenter-video
Turn a topic or finished script into a complete, publish-ready explainer video — led by an AI presenter from an authorized adult presenter image, or performed in one of nine visual explainer styles…
eternityspring/reelbench-skills
拉片:把一条成片拆成逐镜头的分析表——每个镜头的时长、景别、类别、运镜、画面. An agent skill from eternityspring/reelbench-skills.
OSideMedia/higgsfield-ai-prompt-skill
A skill your agent uses whenever the user asks anything about Higgsfield AI — writing or refining video/image prompts, choosing a model (Kling, Veo, Wan, Seedance, Minimax Hailuo, DoP, Soul, Nano…
OSideMedia/higgsfield-ai-prompt-skill
A skill your agent uses when the user asks about Higgsfield Assist (the built-in GPT-5 copilot), how to use the platform's native AI assistant, credit optimization strategies, plan selection, how to…
OSideMedia/higgsfield-ai-prompt-skill
A skill your agent uses when the user wants to generate a cinematic still image on Higgsfield, asks about shot framing, camera angle, or composition for image prompts, needs a specific shot type…
OSideMedia/higgsfield-ai-prompt-skill
A skill your agent uses when the user asks about Mixed Media, wants to apply artistic preset styles to an image (Noir, Sketch, Paper, Canvas, Particles, Neon, etc.), combine multiple artistic…
OSideMedia/higgsfield-ai-prompt-skill
A skill your agent uses when the user asks about Moodboard, building a moodboard from reference images, curated moodboard presets, Soul Hex color transfer, applying a visual style direction to…
OSideMedia/higgsfield-ai-prompt-skill
A skill your agent uses when the user wants to apply a named Higgsfield motion preset, asks about VFX presets, transformation effects, elemental effects, or transition presets.
Categories
A skill your agent uses when the user asks about visual styles, aesthetics, color grades, film looks, or how to set the tone and atmosphere of a Higgsfield generation. Higgsfield Style is an agent skill from OSideMedia/higgsfield-ai-prompt-skill. Use when the user asks about visual styles, aesthetics, color grades, film looks, or how to set the tone and atmosphere of a Higgsfield generation.
Higgsfield Style fits situations like: the user asks about visual styles; how to set the tone and atmosphere of a Higgsfield generation.
Run `npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a claude-code`. Or copy the skill folder (skills/higgsfield-style in OSideMedia/higgsfield-ai-prompt-skill) into .claude/skills/higgsfield-style in your project. Claude Code loads it when a task matches its description.
Run `npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a codex`. Or copy the skill folder (skills/higgsfield-style in OSideMedia/higgsfield-ai-prompt-skill) into .agents/skills/higgsfield-style in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add OSideMedia/higgsfield-ai-prompt-skill --skill higgsfield-style -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/higgsfield-style, .gemini/skills/higgsfield-style, .github/skills/higgsfield-style and .opencode/skills/higgsfield-style in your project.
SKILL.md names no scripts, command-line tools or credentials: Higgsfield Style is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Higgsfield Style is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 3.9k tokens (SKILL.md is roughly 15k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Higgsfield Style: Video Generation (bytedance/deer-flow, 84k stars), Video Cover Image (itwanger/toBeBetterJavaer, 18k stars), Seedance (songguoxs/seedance-prompt-skill, 2.9k stars) and HyperFrames Video Entry Point (heygen-com/hyperframes, 60k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
OSideMedia (a GitHub user) maintains it in OSideMedia/higgsfield-ai-prompt-skill, which has 713 GitHub stars. The repository holds 33 skills in this directory. The repository was last updated on September 27, 2026.
Source: OSideMedia/higgsfield-ai-prompt-skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.