chatcut-workspace

Launches the ChatCut course-video editing workspace and manages clip import, arrangement, and relinking.

Updated Jul 3, 2026
One-click install
npx skills add https://github.com/LYGOLANG/fufan-cc-flow --skill chatcut-workspace-lygolang
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: chatcut-workspace
Source: https://github.com/LYGOLANG/fufan-cc-flow/tree/main/bundled-plugins/chatcut/skills/chatcut-workspace
Command: npx skills add https://github.com/LYGOLANG/fufan-cc-flow --skill chatcut-workspace-lygolang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It removes the friction of starting a course-video editing session: checking heavy dependencies, launching the local ChatCut workbench, importing recordings into the project, and keeping the project consistent when media files move, go missing, or get rearranged. ## Core Features & Use Cases - Workspace startup with dependency gating: Runs dependency_check before opening the workbench, reports blocked features (annotation, ASR, TTS), and walks the user through dependency_fetch with explicit confirmation before downloading large packages like hyperframes or whisper. - Clip lifecycle management: Imports recordings via clip_import, reorders or trims segments with clip_arrange, and relinks missing media with clip_relink while preserving trim points and annotations. - UI request loop handling: Claims pending requests from the workbench queue (narration, captions, export, relink) with request_claim, executes the matching tool with the requestId attached, and reports job progress or cancellation honestly. - Use Case: A user says "import these three screen recordings and move the third clip before the first" — the skill opens the workspace, imports the files, rearranges the timeline, and confirms the new total duration. ## Quick Start Open my ChatCut workspace in the current project directory and import the recordings from my course folder so I can start editing an episode.

Frequently Asked Questions about chatcut-workspace

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I import screen recordings into a ChatCut video project?▼

Open the workspace with workspace_open, then call clip_import with the file paths. The tool probes each file with ffprobe and generates preview proxies without copying or modifying the originals. Files must be inside the project directory, the host working directory tree, or a configured media root.

Why does clip import fail with PERMISSION_DENIED?▼

PERMISSION_DENIED means the file is outside the allowed import scope: the project directory, the host session working tree, user-configured media roots, or previously imported files. Add the directory via the CHATCUT_MEDIA_ROOTS environment variable or the mediaRoots key in the chatcut config.json, then restart the session.

What happens when source video files are moved or deleted?▼

The project stays intact: trim points, annotations, and narration are preserved, and only the media file is missing. Use project_read to list missingClipIds, ask the user for each new path, then call clip_relink per clip. Export is blocked until all clips are relinked.

How do I install missing dependencies like hyperframes or whisper?▼

Call dependency_fetch once without confirm to get a summary of size, source, and install location, present it to the user, then call it again with confirm true after approval. Track progress with job_status and re-run dependency_check afterward to verify the blocked feature cleared.

Why does the workbench UI stay stuck on 'in progress' after clicking a button?▼

UI actions like export or caption generation enter a request queue and are not executed automatically. The agent must call request_claim to take ownership, run the matching tool with the returned requestId, and the queue entry closes when the write completes.

Can dependency_fetch add whisper or libass filters to ffmpeg?▼

No. Whisper and libass filter support in ffmpeg is a compile-time option of the local ffmpeg build, so dependency_fetch cannot supply it. The user must install an ffmpeg build that already includes those filters.