About Ghostie
Ghostie is a cross-platform desktop AI assistant built with Tauri and React, for people who want model-based conversations and agent workflows. It combines configurable agents, local knowledge management, visual workflows, plugins, and desktop controls.
Key features
- 01Connect to multiple AI models
- 02Configure agents and delegate tasks
- 03Build visual workflows
- 04Manage local document knowledge bases
- 05Run local scripts and automate desktop tasks
- 06Extend functionality with plugins
Use cases
- 01Switching between agents for different tasks
- 02Searching imported documents for answers
- 03Automating tasks with workflow triggers
- 04Running local scripts from the desktop assistant
MCP support
Ghostie lists an MCP market for discovering MCP integrations. The README does not describe specific MCP server management or connection behavior.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEghostie Intelligent Assistant
✨ Key Features
🤖 AI & Agent System
- Support for multiple AI models (OpenAI, Tongyi, etc.)
- Multi-agent system with different roles and capabilities
- Intelligent context management and conversation flow
- Customizable agent personalities and behaviors
- Agent collaboration and task delegation
🌐 MCP Support
- MCP market: open with one click
- more from other source
💻 Desktop Integration
- Full desktop system access and control
- Local script execution without MCP server
- System-level automation capabilities
- Native file system integration
- Cross-platform support (Windows, macOS, Linux, Current in Windows)
🔌 Plugin Architecture
- Extensible plugin system
- Custom plugin development support
- Plugin marketplace
- Plugin hot-reload capability
- Plugin permission management
⚡ Workflow Engine
- Visual workflow designer
- Multi-agent workflow orchestration
- Conditional branching and loops
- Error handling and recovery
- Workflow templates and sharing
📚 Knowledge Management
- Local knowledge base support
- Document parsing and indexing
- Multi-format support (PDF, Word, Markdown, TXT)
- Knowledge base search and filtering
- Knowledge sharing and collaboration
🎯 User Experience
- Quick shortcut activation (Alt + Space)
- Modern UI based on Tailwind CSS
- Automatic updates
- Cross-platform consistency
- Full-stack TypeScript development
📸 Interface Preview
🚀 Quick Start
System Requirements
- Windows 10 or above
- macOS or Linux (coming soon)
📖 User Guide
Basic Usage
-
Quick Launch
- Use shortcut
Alt + Spaceto summon the assistant window anytime- Window appears elegantly in the center of the screen
- Supports automatic window position memory
- Customizable shortcut combinations in settings
- Use shortcut
-
Conversation Interaction
- Input Features
- Multi-line input support (Shift + Enter for new line)
- Automatic code block recognition and formatting
Ctrl + Lshortcut to clear input- Up/down arrow keys for quick history access
- Conversation Features
- Markdown format display support
- One-click code block copying
- Context Management
- Automatic conversation context maintenance
- Manual context clearing (
Ctrl + R) - Fixed context settings support
- Smart context length adjustment
- Input Features
-
Assistant Switching
- Preset Assistant Types
- General Assistant: Daily conversations and Q&A
- Code Assistant: Programming-related queries
- Writing Assistant: Content creation
- Analysis Assistant: Data analysis and processing
- Custom Assistants
- Create in Settings > Assistant Management
- Import/export assistant configurations
- Clone existing assistants
- Custom assistant icons and styles
- Preset Assistant Types
Advanced Features
-
Knowledge Base Management
- Document Support
- Supported formats: PDF, Word, Markdown, TXT
- Automatic document parsing and indexing
- Batch folder import support
- Automatic document synchronization
- Knowledge Base Operations
- Create multiple knowledge base groups
- Knowledge base search and filtering
- Knowledge base merging support
- Import/export functionality
- Q&A Features
- Precise Q&A based on knowledge base
- Multi-knowledge base joint query support
- Adjustable relevance threshold
- Answer source tracing
- Document Support
-
Workflow Automation
- Workflow Creation
- Visual workflow editor
- Conditional branching and loops
- Multiple action node combinations
- Error handling and retry mechanisms
- Trigger Conditions
- Scheduled triggers
- Event triggers
- Message triggers
- API triggers
- Execution Control
- Parallel/sequential execution control
- Execution status monitoring
- Log recording and tracking
- Execution result notifications
- Workflow Creation
-
History and Export
- History Management
- View by date/assistant category
- Full-text search support
- Tag management system
- Important conversation starring
- Export Features
- Markdown format export
- HTML webpage export
- PDF document export
- Batch export functionality
- History Management
💡 Core Features
ChatModel
- Support for multiple model integrations (OpenAI, Azure, etc.)
- Intelligent context management
- Custom model parameter configuration
- Complete conversation history
- Streaming output support
Agent/Agent/Proxy/Assistant
- Configurable conversation strategies
- Multi-assistant switching
- Workflow and plugin support
- Knowledge base integration
- Assistant personalization settings
Workflow
- Custom workflow support
- Task chain processing
- Multi-agent collaboration
- Extensible plugin system
MCP (Multi-Cloud Platform)
- 地理位置服务(地址解析、地点搜索、天气查询)
- 路线规划与导航服务(驾车、步行、公交、骑行)
- IP定位服务
- 距离测量与周边搜索
- 可扩展的API集成框架
⚙️ Development Setup
# Install dependencies
npm install
# Run in development mode
npm run tauri dev
# Build application
npm run build:tauri
🤝 Contributing
- Fork this repository
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
📄 License
This project is licensed under the MIT License - see the LICENSE file for details.
🙏 Acknowledgments
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.