Minimax H3

Write, debug and structure prompts for MiniMax H3 video generation (T2VA, I2VA, FL2VA, L2VA, Ref2VA) and configure its ComfyUI workflow. Use when the user mentions MiniMax H3, minimax_h3, fl2va, ref2va, MiniMaxH3ReferenceToVideo, reference-to-video, asks to animate a photo, write a video prompt, pick a model quant, or rebuild a reference clip or reel into a prompt, or wants to fix drifting identity, phantom objects, mangled hands, weightless falls and broken camera moves in H3 output.

16 starsImages, Video & Audio

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add teskor-hub/minimax-h3-skill --skill minimax-h3

Manual install

git clone --depth 1 https://github.com/teskor-hub/minimax-h3-skill.git /tmp/minimax-h3-skill
mkdir -p ~/.claude/skills
cp -r /tmp/minimax-h3-skill/. ~/.claude/skills/minimax-h3

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# MiniMax H3 — prompting and ComfyUI setup

MiniMax H3 is an open-weight omni-modal video model: it generates video **and native stereo audio in a single forward pass**, at 24 fps, with a trained clip length of roughly 5–15 s. It ships as two checkpoints — `fl2va` (frame-conditioned) and `ref2va` (reference-conditioned) — which are different weights, not modes of one model.

This skill follows MiniMax's own prompt-writing guides and adds the failure modes those guides do not cover.

- `references/prompting.md` — the official output format for T2VA / I2VA / FL2VA / L2VA
- `references/reference-mode.md` — the official six-section format for full reference (Ref2VA)
- `references/templates.md` — fill-in templates for every mode
- `references/troubleshooting.md` — symptom → cause → fix, from real failures
- `references/comfyui.md` — checkpoints, quants, VRAM, node-by-node settings
- `references/reel-to-prompt.md` — rebuilding a reference clip: measure its cuts, read its frames, write the prompt from what is there
- `references/reel-modes.md` — ControlNet / Motion Strip / Hybrid selection, source dialogue and mode-specific readiness

Read the full skill on GitHub

Details

Skill name: minimax-h3
Repository: teskor-hub/minimax-h3-skill
License: MIT
Repo updated: 19 September 2026

FAQ

What does the Minimax H3 skill do?

How do I install the Minimax H3 skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add teskor-hub/minimax-h3-skill --skill minimax-h3. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Minimax H3 popular?

It lives in teskor-hub/minimax-h3-skill, which has 16 GitHub stars and 0 forks.

Related skills

Motion Graphics

kajisho5/motion-graphics-skill

Deterministic motion-graphics rendering execution Skill for the AI Video Production Ecosystem. Use it when a caller (normally video-production-agent) has already decided what graphic to show, when to show it, and roughly how it should look, and needs it rendered safely: title cards, lower thirds, free-form text overlays, image/logo overlays, video-on-video picture-in-picture/chroma-key overlays, persistent corner "bug" watermarks, chapter chips, a bottom progress bar, and a countdown, with built-in template animation (title/lower-third/bug/chapter/countdown fade, slide, or pulse) or a configur

Images, Video & Audio

Asd Ste100

prithivrajmu/asd-ste100

Explain or rewrite technical text in clear, STE-inspired English. Use for ASD-STE100, Simplified Technical English, '80% STE', 'strict STE', 'Karpathy-style', or a plain explanation of code, a diff, or a system. Supports diagrams, interactive HTML, and explainer videos when requested. Not for creative or marketing copy.

Images, Video & Audio

Feishu Card JSON V2

lageev/feishu_msg_card_skill

Construct, refine, review, validate, and package polished Feishu/Lark Card JSON 2.0 payloads. Use when creating cards for Feishu custom-bot webhooks, application bots, notifications, alerts, reports, dashboards, approvals, forms, callbacks, updates, streaming AI output, Markdown-rich content, tables, charts, images, audio, localization, or responsive layouts; also use when converting a content brief or legacy card into strict JSON 2.0 or diagnosing a rejected card payload.

Images, Video & Audio

Soul Aaron

aaronjmars/soul-aaronjmars

Embody Aaron's digital identity. Read SOUL.md first, then STYLE.md, then examples/. Become Aaron—opinions, voice, worldview, the works.

Images, Video & Audio

Motion Video Workflow

isaachang/motion-video-workflow

用户提供配音（背景音乐可选，没有就自动合成），要做口播动效视频（案例分享、AI资讯、AI知识讲解、教程）时使用：听写、抓素材、分镜、代码动效，按需出16:9或9:16成片，并出3:4和4:3封面。

Images, Video & Audio

Music Producer

AEON-7/aeon-music-maker

Generate standalone music tracks (songs, instrumentals, cues) at maximum audio fidelity using ACE Step 1.5 XL on the Workstation workstation. USE THIS SKILL when the user asks to make a music track, song, instrumental, beat, album cut, demo, jingle, score cue, or any standalone audio deliverable where music quality is the primary concern (not background bed for a radio drama or video). Triggers on: "make a song", "generate music", "create a track", "music in the style of X", "instrumental", "lofi / jazz / classical / ambient / rock / etc. track", "album", "demo", "jingle", "music cue", "origin

Images, Video & Audio

Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.