A desktop video editing workflow application that streamlines video ingest, project creation, and integrates with professional video production tools.
- ➤ Overview
- ➤ Key Features
- ➤ Installation
- ➤ Ollama Setup
- ➤ How It Works
- ➤ Tech Stack
- ➤ Premiere Pro Plugin Management
- ➤ License
Bucket is a powerful desktop application built with Tauri (Rust + React/TypeScript) designed to streamline video editing workflows for professionals. It simplifies video file ingest, automates project creation, and seamlessly integrates with industry-standard tools like Adobe Premiere, Trello, and Sprout Video.
- AI Script Formatter: AI-powered autocue script formatting using locally hosted LLM models
- Upload Word documents (.docx) and format them for teleprompter use
- Text editor to review and update AI changes
- Edit and refine AI suggestions before exporting
- Powered by local Ollama models (no cloud required)
- Multi-Camera Project Setup: Organize footage by camera assignment for multi-camera shoots
- Adobe Premiere Integration: Automatically generate project templates and folder structures
- Progress Tracking: Real-time progress bars for file operations and project creation
- External Tool Integration:
- Trello: Project management and card updates
- Sprout Video: Video hosting with custom posterframe generation
- Secure User Management: Login/registration with encrypted data storage
- Cross-Platform: Available for Windows, macOS, and Linux
- Bun (package manager)
- Rust (for development)
- Ollama (for AI Script Formatter feature) - see Ollama Setup below
-
Clone the repository:
git clone https://github.com/twentynineteen/bucket.git cd bucket -
Install dependencies:
bun install
-
Build the application:
bun run build:tauri
-
On macOS, open the DMG file in
/target/build/dmgand copy the app to your Applications folder.
To run in development mode:
bun run dev:tauriThe AI Script Formatter feature requires Ollama to be installed and running locally.
-
Download and install Ollama from ollama.com
- macOS: Download the
.dmginstaller - Linux: Run
curl -fsSL https://ollama.com/install.sh | sh - Windows: Download from the Ollama website
- macOS: Download the
-
Verify installation:
ollama --version
Ollama runs as a background service. To start it:
ollama serveThe service runs on http://localhost:11434 by default.
Before using the Script Formatter, download the following language models:
ollama pull llama3.1:latest
ollama pull nomic-embed-text:latest
ollama listModel Selection Tips:
- llama3.2: Best for quick formatting on limited hardware
- Launch Bucket and navigate to Settings
- Find the Ollama URL field (default:
http://localhost:11434) - Update the URL if needed and click Save
- Click Test Connection to verify Ollama is running and see how many models are available
- Navigate to AI Tools > Script Formatter to start formatting scripts
Connection Failed: Check if Ollama is running:
curl http://localhost:11434/api/tagsIf not running, start it:
ollama serveNo Models Available: List installed models:
ollama listInstall a model:
ollama pull llama3:latestPort Conflicts: If port 11434 is in use, you can run Ollama on a different port:
OLLAMA_HOST=0.0.0.0:11435 ollama serveThen update the URL in Bucket Settings to http://localhost:11435
- Upload Script: Select a
.docxfile from your computer (up to 1GB) - Select AI Model: Choose from available Ollama models running on your machine
- AI Processing: The script is automatically formatted for autocue/teleprompter readability
- Review Changes: View side-by-side diff showing original vs. AI-formatted text
- Edit & Refine: Make manual adjustments to the AI's suggestions if needed
- Export: Download the formatted script as a
.docxfile ready for teleprompter use
- Select Video Files: Choose footage files from your file system
- Assign Cameras: Organize files by camera number for multi-camera projects
- Configure Project: Set project title and output folder location
- Create Project: Generate organized folder structure with Adobe Premiere integration
- Track Progress: Monitor file operations with real-time progress updates
- Integrate & Upload: Connect with Trello for project management or Sprout Video for hosting
- Frontend: React 18 + TypeScript + TailwindCSS
- Backend: Tauri 2.0 (Rust)
- UI Components: Radix UI + Lucide icons
- State Management: Zustand + TanStack React Query
- Build Tool: Vite
- AI Integration:
- Vercel AI SDK v5 for unified provider interface
- Ollama (local LLM runtime)
- Monaco Editor for diff visualization and editing
- mammoth.js for Word document parsing
- docx for document generation
Bucket includes an integrated plugin manager for Premiere Pro CEP extensions. Install and update plugins with one click - no manual file operations required.
Metadata management panel that syncs with Bucket's project system. View and edit breadcrumbs directly in your Premiere timeline.
Features:
- View and edit project metadata in Premiere Pro
- Sync with Trello cards and Sprout Video
- Quick access to project resources
- Seamless integration with Bucket workflows
Premiere Pro extension for streamlined workflows.
Features:
- (User to specify features)
- Navigate to Upload Content > Premiere Plugin Manager
- Click Install on any plugin
- Restart Premiere Pro
- Access via Window > Extensions
- macOS: Automatic debug mode configuration
- Windows: Silent installation
- Cross-platform: Works identically on both platforms
Licensed under UNLICENSED.
