Not in the top 1,000 by GitHub stars

LIUFelix2004/God-minmax-H3

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

b510e3736cda → b510e3736cda

Compare on GitHub

View on GitHubAdd to my skills

Verified commit b510e3736cda1f6bfab35288d3dd62e1e38679c5

308GitHub stars
3stars in 7 days
1skill in this repository
0commits in 30 days

Heat

4.5
  • Star growth, last 7 daysNot available
  • Star growth, last 24 hoursNot available
  • Rank movement, last 7 daysNot available
  • Commits, last 30 daysNot available
  • Accounts asking about it through MCP, last 7 daysNot available
  • Hot votes this weekNot available

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

Oct 25, 2026Nov 6, 2026308

30-day change not yet available: not enough history.

Individual skills in this repository

How it works

minimax-h3-video

When your agent uses it

用 MiniMax H3(海螺 3.0)生成成片级视频的完整方法论与提示词库。当用户提到「海螺」「Hailuo」「MiniMax H3」「H3 生成视频」「AI 视频提示词」「文生视频/图生视频/首尾帧/全能参考」,或需要把参考片逆向拆解成提示词、要把多段 H3 片段拼成 20–60 秒成片、要保持跨段的角色与风格一致性、要写带运镜与原生音频指令的分镜提示词时,使用本技能。也适用于用户只是发来一段视频问「这是怎么生成的」或「帮我做个 29 秒的片子」。

What it loads into context

Frontmatter

186 tokens

Always available to the agent

SKILL.md

1,625 tokens

When the skill is activated

Reference documents

12,374 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. 这个技能解决什么
  2. 先读硬约束
  3. 核心工作流
  4. 提示词库
  5. 常见失败与对策
  6. 合规

Scripts executed separately

  • scripts/extract_audio.sh
  • scripts/extract_last_frame.sh
  • scripts/stitch.sh

Files

  • LICENSEasset1,069 bytes
  • PUSH.mddoc1,487 bytes
  • README.mddoc7,013 bytes
  • SKILL.mdcore4,943 bytes
  • assets/previews/01-final-bet.webpasset68,606 bytes
  • assets/previews/02-hero-vlog.webpasset87,894 bytes
  • assets/previews/03-night-comic.webpasset112,576 bytes
  • prompts/01-final-bet.mddoc4,133 bytes
  • prompts/02-hero-vlog.mddoc3,856 bytes
  • prompts/03-night-comic.mddoc3,957 bytes
  • prompts/04-cavalry-opening.mddoc2,815 bytes
  • prompts/05-single-take-dialogue.mddoc3,740 bytes
  • references/h3-api.mddoc5,044 bytes
  • references/prompt-grammar.mddoc3,682 bytes
  • references/stitching.mddoc2,973 bytes
  • scripts/extract_audio.shruns347 bytes
  • scripts/extract_last_frame.shruns338 bytes
  • scripts/stitch.shruns1,978 bytes
Read SKILL.md

# MiniMax H3 视频生成

## 这个技能解决什么

H3 单次生成上限 15 秒。所有「30 秒 / 1 分钟成片」的需求,本质上都是
**分段生成 + 一致性接力 + 拼接**三件事。本技能给出这三件事的标准做法,
外加 5 个已验证跑通的分镜提示词模板。

## 先读硬约束

| 项目 | MiniMax-H3 |
|---|---|
| 单次时长 | **4–15 秒,只接受整数** |
| 分辨率 | 768P / 2K,24fps |
View the complete source on GitHub

Skills included

  • minimax-h3-video

Repository facts

License
MIT
Created
Sep 1, 2026
Last pushed
Sep 1, 2026
Forks
33
Open issues
0
Project activity
Slowing

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 6, 2026

Show all 41 signals by file

PUSH.md

  • Network accessline 3
    仓库已经建好,在本地执行下面的命令即可推送到 https://github.com/LIUFelix2004/God-minmax-H3
  • Network accessline 4
    https://github.com/LIUFelix2004/God-minmax-H3
  • Network accessline 11
    git add . git commit -m "feat: MiniMax H3 视频生成技能与提示词库" git remote add origin https://github.com/LIUFelix2004/God-minmax-H3.git
  • Network accessline 12
    git commit -m "feat: MiniMax H3 视频生成技能与提示词库" git remote add origin https://github.com/LIUFelix2004/God-minmax-H3.git git push -u origin main
  • Network accessline 13
    git remote add origin https://github.com/LIUFelix2004/God-minmax-H3.git
  • Network accessline 20
    cd God-minmax-H3 git init -b main git remote add origin https://github.com/LIUFelix2004/God-minmax-H3.git
  • Network accessline 21
    git init -b main git remote add origin https://github.com/LIUFelix2004/God-minmax-H3.git git fetch origin
  • Network accessline 22
    git remote add origin https://github.com/LIUFelix2004/God-minmax-H3.git git fetch origin git reset --soft origin/main        # 保留本地文件,接上远端历史

README.md

  • Network accessline 10
    ![License](https://img.shields.io/badge/License-MIT-yellow.svg)
  • Network accessline 11
    ![Model](https://img.shields.io/badge/MiniMax--H3-4~15s%20%C2%B7%202K%20%C2%B7%2024fps-F5FF60?labelColor=111)
  • Network accessline 12
    ![Prompts](https://img.shields.io/badge/提示词-5%20条-3158E8)
  • Network accessline 117
    git clone https://github.com/LIUFelix2004/God-minmax-H3.git ~/.claude/skills/minimax-h3-video
  • Network accessline 174
    - [MiniMax 开放平台 · 视频生成文档](https://platform.minimaxi.com/docs/guides/video-generation)
  • Network accessline 175
    - [awesome-minimax-h3-prompts](https://github.com/xianyu110/awesome-minimax-h3-prompts) —— 更大的社区提示词合集

references/h3-api.md

  • Network accessline 3
    官方文档:https://platform.minimaxi.com/docs/guides/video-generation
  • Network accessline 29
    [Credential-related source omitted]
  • Access to secretsline 29
    [Credential-related source omitted]
  • Network accessline 30
    [Credential-related source omitted]
  • Access to secretsline 30
    [Credential-related source omitted]
  • Network accessline 31
    BASE_URL = "https://api.minimaxi.com"
  • Network accessline 42
    "ratio": "16:9",     }     r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)
  • Network accessline 43
    }     r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)     r.raise_for_status()
  • Network accessline 44
    r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)
  • Network accessline 58
    "resolution": "2K",     }     r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)
  • Network accessline 59
    }     r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)     r.raise_for_status()
  • Network accessline 60
    r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)
  • Network accessline 75
    "resolution": "2K",     }     r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)
  • Network accessline 76
    }     r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)     r.raise_for_status()
  • Network accessline 77
    r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)
  • Network accessline 90
    content.append({"type": "audio_url", "audio_url": {"url": u}, "role": "reference_audio"})     payload = {"model": MODEL, "content": content, "duration": duration, "resolution": "2K"}     r = requests.post(f"{BASE_URL}/v2/video_generation", 
  • Network accessline 91
    payload = {"model": MODEL, "content": content, "duration": duration, "resolution": "2K"}     r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)     r.raise_for_status()
  • Network accessline 92
    r = requests.post(f"{BASE_URL}/v2/video_generation", headers=headers, json=payload)
  • Network accessline 99
    while True:         time.sleep(10)         r = requests.get(f"{BASE_URL}/v2/query/video_generation/{task_id}", headers=headers)
  • Network accessline 100
    time.sleep(10)         r = requests.get(f"{BASE_URL}/v2/query/video_generation/{task_id}", headers=headers)         r.raise_for_status()
  • Network accessline 101
    r = requests.get(f"{BASE_URL}/v2/query/video_generation/{task_id}", headers=headers)
  • Network accessline 110
    def download(url: str, path: str):     with open(path, "wb") as f:         f.write(requests.get(url).content)
  • Network accessline 112
    f.write(requests.get(url).content)

scripts/extract_audio.sh

  • Executable scriptline 1
    #!/usr/bin/env bash

scripts/extract_last_frame.sh

  • Executable scriptline 1
    #!/usr/bin/env bash

scripts/stitch.sh

  • Executable scriptline 1
    #!/usr/bin/env bash
  • Dangerous shell operationline 47
    rm -f "$TMP"

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Signal confidence: 41 facts

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeCompatibleStandard SKILL.md SKILL.md.claude/skills/<skill-name>/
CodexCompatibleStandard SKILL.md SKILL.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md SKILL.md.opencode/skills/<skill-name>/
  • Native The repository ships a manifest, folder or frontmatter field for this agent.
  • Compatible A standard SKILL.md that this agent can read.
  • Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.