Use when adding new error messages to React, or seeing "unknown error code" warnings.
npx skills add donghaozhang/video-agent-skill --skill "qcut-technical-docs"
Install specific skill from multi-skill repository
# Description
QCut technical documentation and architecture reference. Use when understanding codebase structure, AI video workflows, testing infrastructure, terminal system, or finding specific implementation details.
# SKILL.md
name: qcut-technical-docs
description: QCut technical documentation and architecture reference. Use when understanding codebase structure, AI video workflows, testing infrastructure, terminal system, or finding specific implementation details.
user-invocable: false
QCut Technical Documentation
Reference documentation for QCut's architecture, workflows, and implementation details.
Documentation Index
For full navigation, see docs/technical/README.md.
Quick Reference
Architecture
| Topic | File | Use When |
|---|---|---|
| Source Code Structure | architecture/source-code-structure.md | Finding files, understanding folder layout |
| Terminal Architecture | architecture/terminal-architecture.md | Understanding xterm.js integration |
| Virtual Folder System | architecture/virtual-folder-system.md | Media organization, folder metadata |
AI Video Generation
| Topic | File | Use When |
|---|---|---|
| AI Workflow | ai/workflow.md | Working with AI video generation, FAL.ai |
| Skills System | ai/skills-system.md | How skills work in QCut |
AI Models
| Category | Folder | Models |
|---|---|---|
| Text-to-Video | ai/models/text-to-video/ | Sora 2, Veo 3, Kling, MiniMax, Wan, LTX |
| Image-to-Video | ai/models/image-to-video/ | Runway Gen-3, Kling, Luma, Veo 2 |
| Avatar | ai/models/avatar/ | Hedra, Sync, Hailuo |
| Transcription | ai/models/transcription/ | ElevenLabs Scribe, Whisper |
| Text-to-Image | ai/models/text-to-image/ | Flux, SDXL, Ideogram |
| Image Upscale | ai/models/image-upscale/ | Real-ESRGAN, Clarity |
| Segmentation | ai/models/segmentation/ | SAM, BiRefNet |
| Adjustment Panel | ai/models/adjustment-panel/ | Color grading, filters |
Testing
| Topic | File | Use When |
|---|---|---|
| Test Infrastructure | testing/infrastructure.md | Writing tests, understanding test setup |
| E2E Testing | testing/e2e.md | End-to-end testing with Playwright |
Workflows
| Topic | File | Use When |
|---|---|---|
| Effects Workflow | workflows/effects-sequence.md | Video effects pipeline |
| Drawing Canvas | workflows/drawing-canvas-sequence.md | Canvas drawing implementation |
Guides
| Topic | File | Use When |
|---|---|---|
| Build Commands | guides/build-commands.md | Building, running, deploying |
Architecture Overview
Tech Stack
- Frontend: Vite 7.0.6 + TanStack Router + React 18.3.1
- Desktop: Electron 37.4.0 (100% TypeScript)
- State: Zustand stores
- Video: FFmpeg WebAssembly
- AI: FAL.ai (40+ models)
- Testing: Vitest 3.2.4 (200+ tests)
Key Directories
apps/web/src/
βββ routes/ # TanStack Router pages
βββ components/ # React components
β βββ ui/ # Radix UI primitives (73 files)
β βββ editor/ # Video editor components
βββ stores/ # Zustand state stores
βββ hooks/ # Custom React hooks
βββ lib/ # Utilities and services
β βββ ai-video/ # AI video generation
βββ types/ # TypeScript definitions
electron/
βββ main.ts # Electron main process
βββ preload.ts # IPC bridge
βββ *-handler.ts # IPC handlers (38 total)
How Claude Uses This Skill
This skill is model-invoked only (user-invocable: false). Claude automatically loads relevant sections when:
- Understanding code structure β Loads architecture/source-code-structure.md
- Working with AI video β Loads ai/workflow.md + model docs
- Writing/debugging tests β Loads testing/*.md
- Terminal features β Loads architecture/terminal-architecture.md
- Build/deploy questions β Loads guides/build-commands.md
# Supported AI Coding Agents
This skill is compatible with the SKILL.md standard and works with all major AI coding agents:
Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.