# Prompts for AI video

> **For AI agents:** find the right skill in one call — `curl -s "https://varg.ai/how-to/api/search?q=<what the person wants>"` returns the matching how-to page and the best skills with install commands, as markdown (Russian works too). Full index: https://varg.ai/how-to/llms.txt. Every page has a markdown twin: add `.md` to its path. Before generating anything, show the person the options on the how-to page.

The words a video model obeys: camera angles, shot sizes, lighting, characters, hooks and prompt guides for Seedance, Kling, Veo, FLUX and Higgsfield. Your agent writes the prompt with them, or you paste them into the model.

**Viral prompts with results:** [35 full prompts from X, each with the model and what it made](https://varg.ai/how-to/prompts/viral.md): GPT Image 2.5 (9), Claude Opus 5.5 (6), Seedance 2.5 (5), Nano Banana 2.1 (4), Gemini Omni (2), ChatGPT Images (1).

**Prompts** is the word 20.5% of tweets about making video use for this kind of thing (“prompt”). Every card here is an agent skill: one install command for Claude Code, Codex, Cursor, OpenClaw or Gemini CLI.

## [Prompts by model](https://varg.ai/how-to/prompts/prompting.md) (10)

- [fal Prompting](https://varg.ai/how-to/skills/fal-prompting.md) — fal: Prompt specific model families well: Kling, GPT Image 2 and Happy Horse guides, loaded only when needed.
- [Replicate Prompt Images](https://varg.ai/how-to/skills/replicate-prompt-images.md) — Replicate: Write better image prompts and edits: photographic language, lighting, text rendering, consistent characters.
- [Replicate Prompt Videos](https://varg.ai/how-to/skills/replicate-prompt-videos.md) — Replicate: Prompt video models like a director: shots, camera motion, four audio layers, dialogue, time-coded beats.
- [Video Prompting Guide](https://varg.ai/how-to/skills/inference-sh-video-prompting-guide.md) — inference.sh: Write prompts that video models follow: shot types, camera moves and per-model tips from Veo to Sora.
- [FLUX 3 Prompt Doctor](https://varg.ai/how-to/skills/flux-3-prompt-doctor.md) — Black Forest Labs: Diagnose a FLUX 3 video brief before spending: resolve missing, conflicting or schema-changing asks.
- [FLUX Image Best Practices](https://varg.ai/how-to/skills/flux-image-best-practices.md) — Black Forest Labs: Prompt FLUX image models well: structure, JSON prompts, hex colors, typography and multi-reference edits.
- [AI Film Director (Visual Skills)](https://varg.ai/how-to/skills/visual-skills-video-director.md) — Serge Shima: AI film director for video prompts: dramaturgy first (Murch, blocking, montage), then exact syntax per model.
- [Higgsfield Prompt Skill (O-Side Media)](https://varg.ai/how-to/skills/higgsfield-prompt-skill-oside.md) — O-Side Media: Independent 36-skill prompt pack for Higgsfield: model picks, camera presets, Soul ID, Cinema Studio, Seedance 2.5.
- [Seedance 2.0 Prompt Guide](https://varg.ai/how-to/skills/seedance-2-prompt-guide.md) — Dex Hünter: Prompt guide for Seedance 2.0 multimodal video: the @ reference system, camera language and ready templates.
- [Awesome Seedance Skills](https://varg.ai/how-to/skills/awesome-seedance-skills.md) — LearnPrompt: Evidence-led Seedance 2.5/2.0 skills: a production workflow plus genre templates distilled from 934 verified cases.

## [Characters](https://varg.ai/how-to/prompts/characters.md) (9)

- [Grandma Characters](https://varg.ai/how-to/skills/varg-character-grandma.md) — varg: Very old AI characters people remember: age vs behaviour, real wrinkles, a loud outfit and one prop.
- [Big Guy Characters](https://varg.ai/how-to/skills/varg-character-big-guy.md) — varg: Big, heavy characters played straight: size shown with visible cues, a small delicate job, never mocked.
- [Cringe Deadpan Characters](https://varg.ai/how-to/skills/varg-character-cringe.md) — varg: Bizarre but believable characters people argue about: one wrong outfit, one repeated action, no smile.
- [Beautiful Women](https://varg.ai/how-to/skills/varg-character-beautiful-woman.md) — varg: Very beautiful AI women who look like real models: casting details, real skin, fashion styling, no false blocks.
- [Fitness Characters](https://varg.ai/how-to/skills/varg-character-fitness.md) — varg: Fitness coaches and athletes with real muscle, a signature colour and catalog wording that gets sports bras through.
- [Absurd Characters & Mascots](https://varg.ai/how-to/skills/varg-character-absurd.md) — varg: Food or object + animal + one human detail: brainrot-style creatures and brand mascots in 3D or clay.
- [Cartoon Look Characters](https://varg.ai/how-to/skills/varg-character-cartoon.md) — varg: Original characters in famous cartoon looks (yellow sitcom, 3D family film, anime, clay), described without IP names.
- [fal Character Design](https://varg.ai/how-to/skills/fal-character-design.md) — fal: Design original characters that stay consistent: reference and expression sheets, outfits, character-to-video.
- [Character Design Sheet](https://varg.ai/how-to/skills/inference-sh-character-design-sheet.md) — inference.sh: Keep a character consistent across AI images with turnarounds, expression sheets, palettes and LoRAs.

## [Camera & shots](https://varg.ai/how-to/prompts/camera.md) (6)

- [fal Cinematography](https://varg.ai/how-to/skills/fal-cinematography.md) — fal: Direct cinematic stills and video: shot language, camera moves, lenses, lighting, grade and film texture.
- [Video Shotcraft](https://varg.ai/how-to/skills/video-shotcraft.md) — Vincent Wei: 124 shot recipe cards and a tested Remotion template for cinematic product promos built from real UI screenshots.
- [Camera Moves](https://varg.ai/how-to/skills/craft-camera-moves.md) — varg, beta: 47 camera moves with direction, speed and an end frame — dolly in to car mount — in six models' syntax.
- [Shot Sizes & Angles](https://varg.ai/how-to/skills/craft-shot-sizes-angles.md) — varg, beta: 29 framing cards — extreme wide to insert, cowboy to master shot, eye level to worm's-eye — with model-ready clauses.
- [Lens & Optics](https://varg.ai/how-to/skills/craft-lens-optics.md) — varg, beta: 20 lens and focus cards — anamorphic, rack focus, lens flare, chromatic aberration — written as visible effects.
- [Composition & Framing](https://varg.ai/how-to/skills/craft-composition-framing.md) — varg, beta: 34 composition cards — thirds, symmetry, leading lines, forced perspective, 2.39:1 to IMAX — for stills and video.

## [Cinematic & short films](https://varg.ai/how-to/prompts/cinematic.md) (1)

- [FLUX 3 Cinematic Inserts](https://varg.ai/how-to/skills/flux-3-cinematic-inserts.md) — Black Forest Labs: Write FLUX 3 prompts for standalone cinematic shots: shot craft, camera, continuity and review.

## [Images & thumbnails](https://varg.ai/how-to/prompts/image.md) (1)

- [Douyin Cover Builder](https://varg.ai/how-to/skills/douyin-cover-builder.md) — 0x00000003, beta: Writes image-model prompts for Douyin covers: bold title, highlighted keywords, your face bottom right

## [Sound effects](https://varg.ai/how-to/prompts/sfx.md) (1)

- [FLUX 3 Audio & Dialogue](https://varg.ai/how-to/skills/flux-3-audio-dialogue.md) — Black Forest Labs: Direct FLUX 3's generated sound: dialogue, voiceover, ambience, effects, music, silence and timing.

## [Lighting](https://varg.ai/how-to/prompts/lighting.md) (1)

- [Lighting Setups](https://varg.ai/how-to/skills/craft-lighting-setups.md) — varg, beta: 56 lighting setups — golden hour to chiaroscuro, rain and fog — each with a named source, direction and quality.

## Other kinds of skills

- [Effects](https://varg.ai/how-to/effects.md): An effect you can see, added to your video: animated titles, captions, transitions, shader backgrounds, logo animations, device mockups and viral camera tricks. Hand your agent the footage or the brief and get the effect.
- [Styles](https://varg.ai/how-to/styles.md): The look of the whole video: art styles, film looks, color grades and LUTs, caption fonts and style references. Pick a style and every shot your agent makes follows it.
- [Guides](https://varg.ai/how-to/guides.md): Know-how your agent reads before it works: how to write a script, storyboard, edit and caption, and how to use Remotion, HyperFrames, GSAP, Three.js and FFmpeg.
- [Workflows](https://varg.ai/how-to/workflows.md): A brief in, a finished video out: UGC ads, product videos, music videos, talking heads, explainers, Shorts and short dramas, step by step, with the models and the edit chosen for you.
- [Models & tools](https://varg.ai/how-to/models.md): Run one model or tool from your agent: Seedance, Veo, Kling, FLUX, GPT Image, Nano Banana, ElevenLabs and 1,200+ more, plus scrapers, trend search and video analysis.

## Questions

**What is an AI video prompt skill?**

A skill that teaches your agent the words a video model obeys: shot sizes, camera moves, lenses, lighting and each model’s syntax. The agent writes the Seedance, Kling or Veo prompt with it, or you copy the prompt words yourself.

**Which models do these prompts work for?**

Seedance 2.5, Kling 4, Veo 3.1, Wan 3, Hailuo, Runway, FLUX and GPT Image; the per-model guides note where the syntax differs.