口播视频自动剪辑 MCP
Local checks for talking-head videos, with optional paid cloud rendering via MCP
Answered our latest tool-list check an hour ago with 3 tools. It will be marked verified after 3 successful checks in 7 days.
- Pricing
- Freemium
- Platforms
- CLI, API, macOS
- Runs
- Remotely
- Version
- 1.2.0
- Listing updated
- today
Similar MCP servers
About 口播视频自动剪辑 MCP
口播视频自动剪辑 MCP is a freemium MCP server for content publishing. It runs as a hosted remote server, so there is nothing to install locally.
The README says the core tool is a set of local Python and ffmpeg scripts that inspect a recorded talking-head video before editing. They look at where the face sits so overlay cards avoid it, split subtitles on word boundaries, and verify audio for clipping, gain and timing. Results come back as a JSON or readable report with a PASS or FAIL verdict. Video files stay on the local machine and no account is required for this part.
A separate paid layer renders the edit. Its cloud service takes a plan built from the local checks and produces a finished video, with the first three renders free and later ones priced per render. The same service is also exposed as a remote MCP endpoint with generate, render and job-status tools, authenticated by an API key.
The product suits solo creators and small teams who record spoken-word videos regularly and want to catch layout problems before spending time on editing or outsourcing.
Install
Run in a terminal
claude mcp add --transport http th-recut-mcp https://www.qianshi.cool/thapi/mcpAdd to Settings → Connectors → Add custom connector
https://www.qianshi.cool/thapi/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"th-recut-mcp": {
"url": "https://www.qianshi.cool/thapi/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"th-recut-mcp": {
"type": "http",
"url": "https://www.qianshi.cool/thapi/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"th-recut-mcp": {
"serverUrl": "https://www.qianshi.cool/thapi/mcp"
}
}
}Add to Grok Bot desktop app → Connect apps → Add a custom connector
https://www.qianshi.cool/thapi/mcpEnter a Name (for example "口播视频自动剪辑 MCP"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.
Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"th-recut-mcp": {
"type": "http",
"url": "https://www.qianshi.cool/thapi/mcp"
}
}
}Using several servers? Build one config file with the MCP config generator.
Tools this server provides (3)
Read directly from the server.
- generate_recut_plan
- 根据口播视频元信息与文案,生成出片计划(人脸禁区安全布局 / MG 卡片反选 / BGM 情绪匹配),不渲染。
- render_recut_video
- 提交渲染任务,返回 job_id;用 get_recut_job 轮询成片。
- get_recut_job
- 查询渲染任务状态与成片 URL。
What it does
- Face safe-zone detection
- Works out the area occupied by the face so that overlay cards can be placed outside it, or flagged for manual review.
- Word-boundary sentence splitting
- Splits transcript tokens into segments without cutting through words, producing output meant for a separate editing tool.
- Audio quality checks
- Checks clipping, gain and timing, and flags audio that drifts out of sync with the picture.
- Card and PIP layout validation
- Tests whether card rectangles cover the face zone and verifies picture-in-picture placement and caption alignment.
- Optional cloud rendering
- Renders the finished video from the layout plan through a paid service, with the first three renders free.
Good for
- Check a talking-head recording for face placement and caption breaks before editing
- Verify audio levels and sync on a finished cut without a manual review pass
- Generate a layout plan from local checks and send it to a cloud renderer
- Connect an AI coding assistant to a remote rendering endpoint over MCP
Pricing
Freemium
Local checks are free under CC BY-NC 4.0; cloud rendering has three free renders, then ¥2.90 per render, or monthly subscription tiers on WorkBuddy.
Prices change. Check qianqiuwanzi.github.io before you buy.
口播视频自动剪辑 MCP FAQ
Does the local tool upload my videos?
The README says the scripts only read local files, decode them with the local ffmpeg, and make no network requests.
What are the requirements for the local scripts?
Python 3.8 or later and ffmpeg on the PATH. numpy, opencv and jieba are optional and only needed by the scripts that use them.
Can I use the project commercially?
The README states the repository is licensed CC BY-NC 4.0, which prohibits commercial use, and directs commercial users to the paid cloud rendering service or to the author.
How is the cloud MCP endpoint authenticated?
With an Authorization Bearer header carrying an API key issued by the platform.
Is 口播视频自动剪辑 MCP free?
口播视频自动剪辑 MCP has a free plan, with paid plans for more usage or features.
How do I add 口播视频自动剪辑 MCP to Claude Code?
Run this command in your terminal: claude mcp add --transport http th-recut-mcp https://www.qianshi.cool/thapi/mcp
What are the best alternatives to 口播视频自动剪辑 MCP?
The closest alternatives to 口播视频自动剪辑 MCP are Levea — Autonomous AI Video Editor (37 Tools, 280+ Caption Templates, OAuth 2.1), Clipzi, Premiere MCP, GeekLink Subtitle MCP (Developer Preview) and VideoDB.
Where this comes from
- Description: written from the project's README, the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.