jianying-bridge
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@jianying-bridge Build a new JianYing draft from ~/edit-manifest.json and publish it"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
JianYing Bridge
安全地把一个紧凑的 JSON 剪辑计划编译成可继续编辑的剪映专业版草稿。
A local Codex plugin and MCP server that compiles one compact JSON edit plan into a new, editable JianYing Pro draft.
非剪映官方项目。当前实机验证环境:macOS + 剪映专业版 11.4.2。
为什么做它
很多剪映自动化项目仍输出旧版 draft_content.json,或直接修改用户已有工程。JianYing Bridge 选择了更保守的边界:
只创建新草稿,不覆盖、删除或修补现有草稿;
先在沙盒生成并校验,再显式发布到剪映草稿库;
视频、音频会复制进草稿自身,降低 macOS 权限和源文件移动导致的丢失风险;
一次提交完整 manifest,避免“一个片段一次工具调用”的高频往返;
全程使用本地素材和本机工具,本身不消耗图片/视频生成额度。
Related MCP server: jianying-local-mcp
已验证能力
剪映专业版 11.4.2(macOS)可发现并打开生成的草稿;
视频、BGM、中文字幕可正常进入时间线并播放;
剪映打开后会将草稿迁移为原生
Timelines/project.json布局;同名工程会被拒绝,绝不静默覆盖;
MCP stdio 初始化、工具发现和调用通过;
5 项本地回归测试通过。
MCP 工具
工具 | 用途 |
| 检查剪映、草稿目录、FFmpeg 和安全边界 |
| 在沙盒目录新建可编辑草稿 |
| 检查项目身份、时间线和本地素材引用 |
| 将已校验的新草稿复制到剪映项目库 |
快速开始
依赖:macOS、剪映专业版 11.4.x、Python 3.11+、uv、FFmpeg/FFprobe。
git clone https://github.com/DeathKnight88/jianying-bridge.git
cd jianying-bridge
./scripts/launch_jianying_bridge若作为 Codex 插件使用,可把仓库目录注册为本地插件源;.mcp.json 已配置好 stdio 服务入口。新安装的插件需要在新的 Codex 任务中加载。
Manifest 示例
{
"version": 1,
"project": {
"name": "My_New_Project",
"width": 1920,
"height": 1080,
"fps": 30
},
"video": [
{
"path": "/absolute/path/clip.mp4",
"start": 0,
"source_start": 0,
"duration": 3,
"track": "VideoTrack",
"transition": {"name": "叠化", "duration": 0.3}
}
],
"audio": [
{
"path": "/absolute/path/music.mp3",
"start": 0,
"duration": 3,
"track": "BGM",
"volume": 0.2
}
],
"text": [
{
"content": "字幕",
"start": 0,
"duration": 3,
"track": "Subtitles"
}
]
}时间单位统一为秒。相对素材路径以 manifest 所在目录为基准。完整模板见 assets/example-manifest.json。
推荐工作流
本地素材
→ FFprobe / Whisper / BaoCut / auto-editor(可选)
→ 一份 edit manifest
→ JianYing Bridge 沙盒生成 + 校验
→ 发布到剪映
→ 人工预览与导出桥接器只负责执行已经确定的剪辑计划。转写、脚本判断、镜头选择等工作建议在生成 manifest 前一次性完成。
当前限制
v0.1 仅支持新建草稿,不编辑剪映 11.x 已有加密工程;
发布时要求剪映主进程关闭,避免项目索引并发写入;
最终预览和导出仍需在剪映中完成;
当前只在 macOS 剪映专业版 11.4.2 做过端到端验证;Windows 和其他版本欢迎测试反馈;
剪映内部格式属于非公开实现,版本升级可能导致兼容性变化。
开发与验证
uv run --with pytest pytest -q tests
uv run --with pyyaml python /path/to/plugin-creator/scripts/validate_plugin.py .第三方代码
运行时包含经过固定版本的 luoluoluo22/jianying-editor-skill 与其内嵌的 pyJianYingDraft。许可证和修改说明见 THIRD_PARTY_NOTICES.md。
License
项目自身代码使用 MIT License。第三方代码继续遵循各自的 MIT 或 Apache-2.0 许可证。
“剪映”及相关商标归其权利人所有。本项目与字节跳动或剪映官方无隶属、授权或背书关系。
Available Tools
4 toolsjianying_build_projectB
Compile one version-1 edit manifest into a NEW editable draft in a sandbox output root.
Inputs: output_root (required), manifest_path (required).
jianying_doctorC
Check JianYing Pro, draft-root, FFmpeg, and the create-new safety boundary.
Inputs: drafts_root.
jianying_publish_projectA
Copy a validated NEW draft into JianYing's library. Refuses overwrite and requires JianYing closed.
Inputs: project_path (required), drafts_root.
jianying_validate_projectC
Validate draft identity, timeline, materials, and local file references.
Inputs: project_path (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool targets a distinct stage: environment check, build, validate, and publish. No overlap in purpose or output.
All names use the consistent prefix jianying_ followed by a clear action (doctor, build, validate, publish). Style is uniform and predictable.
Four tools form a tight, well-scoped lifecycle for the server's purpose: checking prerequisites, building, validating, and publishing drafts. No redundancy or bloat.
The domain is a build-and-publish bridge for JianYing drafts. The four tools cover the full pipeline from environment check to final publication, with no obvious missing operations.
Maintenance
Related MCP Connectors
Render video and run AI media tasks from a single declarative JSON request.
AI editor to build, animate & export layered short-form video projects via one tool catalog.
Render an MP4 from your own images, clips, voice-over and timed captions. JSON to video.
CapCut MCP server for ChatGPT, Claude and any AI agent: list drafts, export draft, add auto captions, share draft for review, invite members to a space, set member role, transfer editing permission and search templates in your own CapCut account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to CapCut yourself; Cotlet does not store your password and you need no CapCut API key. About the official CapCut API: Partly. CapCut has an official plugin for the ChatGPT desktop app that builds drafts and adds captions, effects and templates (capcut.com, checked 6 Nov 2026). Setup and FAQ: https://cotlet.dev/integrations/capcut-api-mcp-server
Related MCP Servers
- FlicenseAqualityDmaintenanceCreates and edits CapCut desktop drafts locally by writing project files, enabling timeline manipulation through MCP tools.7-
- AlicenseBqualityBmaintenanceEnables AI to create and edit local Jianying/CapCut draft projects on macOS from videos, images, audio, and text, supporting editable subtitles, keyframes, motion presets, and batch operations via MCP tools.2213MIT
- AlicenseAqualityCmaintenanceEnables AI agents to programmatically edit CapCut Desktop and JianYing Pro projects by directly manipulating native draft files, supporting multi-track timelines, keyframing, transitions, subtitles, and headless previews.244MIT
- AlicenseNot gradedqualityCmaintenanceEnables MCP clients to build CapCut video draft projects on disk — adding video and image clips, audio with fades, text overlays, SRT subtitles, filters, transitions, animations and keyframes, then saving the draft for editing and opening in CapCut. It also lists existing drafts, browses available effects and fonts, and summarizes track, duration and segment details.MIT