# MiniMax-AI/MiniMax-H3

Repository: https://github.com/MiniMax-AI/MiniMax-H3
Canonical: https://ross.abutalabs.com/products/minimax-h3
Language: Python
License Family: other
Last push: 2026-08-15T08:31:17+00:00

## Health v2 (maintenance only)
Score: 56/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 35, longevity 2
- inputs: {"age_days": 34, "days_push": 18, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7161, forks 470 (observed 2026-08-28T04:09:56.228731+00:00)

## What it is
MiniMax H3 is a repository of installable agent skills for the MiniMax H3 video generation model, including a prompt-writing skill with reference guides for text/keyframe and full-reference (Ref2VA) modes plus eight style-specific video generation skills. Skills are plain Markdown packages installable via npx and compatible with any agent harness that can read a SKILL.md.

## Use cases
- write better prompts for MiniMax H3 video generation
- install a prompt-writing skill into Claude Code or Cursor
- generate videos in specific styles with prebuilt agent skills
- learn prompt patterns for keyframe and reference-based video modes
- use video generation skills with OpenAI Codex or LangChain agents

## When to choose
- you generate videos with the MiniMax H3 model or Hailuo AI and want guided prompting
- you want drop-in agent skills that work across Claude Code, Cursor, Windsurf, or Codex
- you need reference guides for text-to-video and reference-to-video prompting

## When to avoid
- you use video generation models from other providers without H3-compatible prompting
- you need a runtime library or API client rather than prompt-writing guidance
- you require a permissively licensed codebase and cannot accept the model's license terms

## Facets
- artifact type: plugin
- maturity: active
- function: prompt-engineering, video-processing, agent-framework
- domain: media, developer-tools
- platform: cross-platform, cli
- tags: agent-skills, video-generation, prompt-writing, minimax, hailuo-ai, text-to-video, skill-package, video, ai-agents

## Member repositories
- MiniMax-AI/MiniMax-H3 (main) score 56

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:56.228731+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:39:58.212972+00:00, confidence not recorded.
  - readme: https://github.com/MiniMax-AI/MiniMax-H3 (fetched 2026-08-28T04:09:56.228731+00:00, sha 43f179b1cc96)
- Data as of 2026-08-30T08:39:29.467469+00:00.