By T0UGH
Automate video production from storyboarding to publishing using the videoclaw CLI, with AI-powered image/video generation, TTS audio, and direct uploads to Douyin, Kuaishou, and Google Drive.
Use when user needs to generate TTS voice, sound effects, and background music for a video
Use when user needs to view, get, or set project or global configuration
Use when user needs to transform or edit images using AI. Independent image-to-image command for converting reference images to different styles or content.
Use when user needs to convert storyboard images into video clips (requires storyboard step completed)
Use when user needs to convert images into video clips with custom prompts
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
English | AI 视频创作 CLI 工具 | 宿主中立 Skill Pack + CLI
videoclaw 是 AI 视频创作 CLI 工具,目标是把视频创作拆成:
当前推荐组合是:
codex-host-image(如果你有 Codex / ChatGPT 登录环境)也就是说,videoclaw 现在优先把“有 Codex 订阅就能直接出图”作为默认体验路径,同时保留其他正式 provider 和宿主中立演进方向。
uvx videoclaw --help
pip install videoclaw
git clone https://github.com/T0UGH/videoclaw.git
cd videoclaw
pip install -e .
export ARK_API_KEY=your-ark-api-key
export DASHSCOPE_API_KEY=your-api-key
export GOOGLE_API_KEY=your-api-key
export OPENAI_API_KEY=your-api-key
videoclaw 的分发现在分成两层:
uvx / pip)Claude Code 目前仍可通过 marketplace 安装适配层:
/claude install marketplace https://github.com/T0UGH/videoclaw/raw/main/.claude-plugin/marketplace.json
这个 marketplace 入口现在应理解为 Claude Code adapter,而不是 videoclaw 的唯一分发中心。
长期方向是让 skills/ 成为 canonical skill pack,再为不同宿主提供薄 adapter:
# 初始化一个 project
videoclaw init my-project
# 在 project 下创建一个视频单元
videoclaw video create my-project demo-video
# 查看视频列表
videoclaw video list my-project
videoclaw 现在采用:
project 承载共享资产与多个视频;video 是独立产出单元;render/ 管理生成过程;clips/。推荐目录结构:
my-project/
├── .videoclaw/
│ ├── config.yaml
│ ├── index.json
│ └── logs/
├── assets/
│ ├── characters/
│ ├── scenes/
│ ├── props/
│ └── covers/
├── videos/
│ └── demo-video/
│ ├── meta.json
│ ├── brief.md
│ ├── storyboard/
│ ├── images/
│ ├── clips/
│ └── audio/
└── exports/
图片能力正在升级为双轨模型:
openai-image:正式 provider(长期方向)codex-host-image:宿主适配器(低门槛接入)当前配置推荐统一使用:
videoclaw config --project my-project --set models.image.backend=codex-host-image
videoclaw config --project my-project --set models.image.auth=chatgpt_login
videoclaw config --project my-project --set models.image.codex_mode=exec
如果你不走 Codex 订阅链路,再切到其他 provider,例如:
videoclaw config --project my-project --set models.image.backend=gemini
videoclaw config --project my-project --set models.image.model=gemini-3-pro-image-preview
也兼容新的 backend 命名:
videoclaw config --project my-project --set models.image.backend=openai-image
videoclaw config --project my-project --set models.image.auth=api_key
如果你配置了 OPENAI_API_KEY,openai-image 现在会直接调用 OpenAI 官方图片 API,而不是再走过渡兜底。
单段视频会在 videos/<slug>/render/ 下落盘:
input/reference.jsoninput/prompt.mdcandidates/selected.mp4task.jsonvideoclaw video create my-project demo-video
videoclaw video list my-project
videoclaw video status my-project demo-video
videoclaw video generate my-project demo-video --prompt "walk forward" --provider mock
videoclaw video select my-project demo-video v001.mp4
| 提供商 / Backend | 图像 | 视频 | 音频 |
|---|---|---|---|
| volcengine | Seedream | Seedance 2.0 | TTS |
| dashscope | wan2.6-t2i | wan2.6-i2v | cosyvoice-v2 |
| gemini | Nano Banana Pro | - | - |
| openai-image | 规划中 / 接入中 | - | - |
| codex-host-image | 宿主适配路径 | - | - |
| mock | 测试用 | 测试用 | 测试用 |
所有视频创作流程通过 skills/ 目录中的 skill pack 定义。当前主要 skill 包括:
video-quick-createvideo-text-storyboardvideo-t2ivideo-i2ivideo-i2vvideo-audiovideo-mergevideo-configvideo-uploadvideo-publish-douyinvideo-publish-kuaishou| 命令 | 说明 |
|---|---|
videoclaw init | 初始化 project |
videoclaw video create | 创建 video |
videoclaw video list | 列出 video |
videoclaw video status | 查看 video 状态 |
videoclaw video generate | 生成单段视频候选 |
videoclaw video select | 选择单段视频候选 |
videoclaw t2i | 文生图 |
videoclaw i2i | 图生图 |
videoclaw i2v | 图生视频 |
videoclaw audio | 生成音频 |
videoclaw merge | 合并视频 |
videoclaw config | 配置管理 |
videoclaw upload | 云盘上传 |
videoclaw preview | 预览文件 |
videoclaw publish | 发布到社交平台(可选依赖) |
详细配置项清单见 docs/configuration.md
配置优先级:
~/.videoclaw/config.yaml<project>/.videoclaw/config.yamlpytest
ruff check .
black .
MIT
npx claudepluginhub t0ugh/videoclawKling AI skill pack - 30 skills for AI video generation, image-to-video, text-to-video, and production workflows
AI video generation — describe what you want, Pexo picks the best model across 10+ engines (Seedance, Kling, Veo, Sora) and returns a finished, multi-shot video with music, subtitles, and transitions. Includes the Pexo agent plus image, audio, director, and model-prompting skills.
Full video production pipeline for Remotion — gives Claude eyes (video analysis), voice (TTS/voiceover), ears (music/SFX), stock footage, AI image/video generation, TikTok captions, 3D content, and more. By Dojo Coding Labs.
Transform text scripts into professional videos with AI voiceovers, stock footage, and Remotion rendering. Advanced features include direct filesystem access, environment management, and whitelisted command execution.
Pika Creative Suite — give any AI agent a face, a voice, and a full creative studio. 9 curated /pika:* skills backed by 58 atomic media tools at mcp.pika.me.
The creative suite for AI agents — motion graphics, asset generation, and video rendering via Remotion + @oanim/core