Flock

Flock

English | [简体中文](docs/README_zh.md)

A multi AI agent desktop application built with Rust and Tauri.

> Note: This project is built on top of langgraph-rust, which is my personal Rust implementation of the LangGraph framework.

> Refactoring History: Flock has been completely rewritten from the ground up. The original version was a Python-based application using LangGraph, LangChain, and FastAPI as the backend. The current version is a native desktop application with a Rust backend, powered by Tauri for the desktop shell. This rewrite brings significant improvements in performance, reliability, and user experience.

> Legacy Code: The original Python codebase is preserved in the legacy/python branch for reference.

Overview

Flock is a desktop application that provides an interactive interface for AI agents with tool orchestration capabilities. It supports multiple LLM providers and features a rich set of built-in tools, skills system, and memory management.

Features

• Multi-Provider Support: OpenAI-compatible, Anthropic, AWS Bedrock, Google Vertex
• Tool Orchestration: Built-in tools (Read, Write, Edit, Bash, Grep, Glob) + MCP server integration
• Skills System: Reusable prompt templates with YAML frontmatter, hot-reload support
• Memory System: Persistent cross-session memory (user, feedback, project, reference types)
• Session Management: Conversation history with SQLite-backed persistence
• Desktop UI: Modern interface built with React, TypeScript, and Mantine UI
• Internationalization: Multi-language support (Chinese/English)

Architecture

Rust Backend

| Crate | Purpose | |-------|---------| | flock-core | Core types, configuration, database, IPC interface, cryptography | | flock-agent | Agent engine, session management, tool execution, memory, graph orchestration | | flock-tools | Tool registry, built-in tools, MCP integration, math/weather tools | | flock-skills | Skill discovery, loading, frontmatter parsing, hooks, permissions | | flock-ui/src-tauri | Tauri desktop app backend |

Frontend Stack

• React 18 + TypeScript
• Mantine UI v8
• Zustand (state management)
• React Query (server state)
• react-markdown + react-syntax-highlighter
• i18next (internationalization)

Getting Started

Prerequisites

• Rust 1.77.2+
• Node.js 18+
• npm or yarn

Dependencies

This project relies on langgraph-rs, which is automatically resolved as a Git dependency in Cargo.toml.

Build & Run

# Clone the repository
git clone https://github.com/Onelevenvy/flock.git
cd flock

# Install frontend dependencies
cd flock-ui
npm install
cd ..

# Build and run the desktop app
cd flock-ui
npm run tauri dev

Development

# Build Rust backend
cargo build

# Run tests
cargo test

# Lint
cargo clippy --workspace

# Frontend development
cd flock-ui
npm run dev
npm run lint

LangGraph Integration

This project leverages langgraph-rust for agent graph orchestration. The LangGraph framework provides:

• State management for agent conversations
• Tool execution flow control
• Checkpointing for conversation persistence
• Pre-built agent patterns

While langgraph-rust is not the official LangGraph Rust implementation, it provides the core features needed for building AI agents with tool orchestration capabilities.

Roadmap

• [ ] Workflow: Visual workflow builder for complex agent orchestration
• [ ] Multi-Agent: Support for multiple agents collaborating on tasks
• [x] Scheduled Tasks: Automated task execution with cron-like scheduling
• [ ] Sandbox: Isolated execution environment for code and commands
• [ ] Browser Tools: Web browsing and interaction capabilities for agents
• [ ] Extensions: Integration with Claude Code, OpenCode, OpenClaw, Hermes, and other third-party agents

License

This project is licensed under the Apache License, Version 2.0 - see the [LICENSE](LICENSE) file for details.

Acknowledgments

• LangGraph - The original Python framework that inspired langgraph-rust
• Tauri - Desktop application framework
• Mantine - React UI component library

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Contact

For questions or feedback, please open an issue on GitHub.

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: ferris-joy
• Source: ferris-joy/flock
• License: Apache-2.0

Install and usage instructions live in the source repository linked above.