OpenFang

v1.0.0Business Applicationsstable

Open-source Agent Operating System

agent-frameworkai-agentsllmmcpopen-source
Share:
17,609
Stars
0
Downloads
0
Weekly
0/5

What is OpenFang?

OpenFang is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to open-source agent operating system

Open-source Agent Operating System

This server falls under the Business Applications category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Open-source Agent Operating System

Use Cases

Agent operating system
Open-source framework
Autonomous agent deployment
RightNow-AI

Maintainer

LicenseApache-2.0
Languagerust
Versionv1.0.0
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx openfang

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use OpenFang

OpenFang is an open-source Agent Operating System built in Rust that lets you run persistent, autonomous AI agents capable of handling web automation, research, social media, lead generation, OSINT, and forecasting tasks. It exposes 53 tools and an MCP-compatible interface alongside an OpenAI-compatible REST API, and includes 40 messaging adapters and 7 autonomous 'Hands' — specialized agent roles that run continuously on schedules. Developers and power users deploy OpenFang when they need a self-hosted, production-grade agent runtime that goes beyond one-shot LLM queries, supporting multi-step workflows, credential vaults, and long-running background agents.

Prerequisites

  • A Linux, macOS, or Windows system with curl available for the installer
  • An LLM API key (OpenAI, Anthropic, or compatible provider) for agent reasoning
  • An MCP-compatible client such as Claude Code or Cursor (for MCP integration)
  • At minimum 2 GB RAM for running the OpenFang binary
  • Optional: messaging platform credentials for WhatsApp, Slack, or other adapters
1

Install OpenFang via the official installer

Download and run the OpenFang installer script. This fetches the latest binary for your platform and places it in your PATH.

curl -fsSL https://openfang.sh/install | sh
2

Initialize your OpenFang workspace

Run the init command to create the default configuration directory and generate a starter config file with credential vault setup.

openfang init
3

Configure your LLM provider

Edit the generated config to point at your LLM provider and API key. OpenFang uses an AES-256-GCM credential vault to store secrets securely.

export OPENFANG_URL=http://127.0.0.1:4200
export OPENFANG_DEFAULT_AGENT=assistant
4

Start the OpenFang server

Launch the OpenFang runtime. It starts the local API server on port 4200 and activates configured agents. The MCP endpoint and OpenAI-compatible API are both served from this process.

openfang start
5

Activate an agent Hand and begin chatting

Activate one of the built-in autonomous Hands (e.g., 'researcher') and interact with it via the CLI or your MCP client.

openfang hand activate researcher
openfang chat researcher

OpenFang Examples

Client configuration

MCP client configuration to connect to the locally running OpenFang MCP endpoint. OpenFang must be started with 'openfang start' before connecting.

{
  "mcpServers": {
    "openfang": {
      "url": "http://127.0.0.1:4200/mcp"
    }
  }
}

Prompts to try

Example prompts to use with OpenFang's researcher and assistant agents via MCP or the CLI chat interface.

- "What are the emerging trends in AI agent frameworks?"
- "Research the top 5 open-source MCP servers and summarize their capabilities"
- "Find recent news about Rust-based infrastructure tools"
- "Generate a lead list of AI startups founded in 2024"
- "Monitor this GitHub repo and alert me to new releases"

Troubleshooting OpenFang

openfang start fails with 'address already in use' on port 4200

Another process is using port 4200. Set the OPENFANG_URL environment variable to a different port (e.g., http://127.0.0.1:4201) and restart.

Agents respond but cannot access the web or external tools

Check that your LLM API key is correctly configured in the credential vault. Run 'openfang init' again if the vault was not set up, and verify network connectivity from the server host.

WhatsApp gateway not connecting

Ensure WHATSAPP_WEB_GATEWAY_URL and WHATSAPP_GATEWAY_PORT environment variables are set. The gateway requires a separately running WhatsApp Web session — consult the OpenFang docs at openfang.sh/docs for gateway setup.

Frequently Asked Questions about OpenFang

What is OpenFang?

OpenFang is a Model Context Protocol (MCP) server that open-source agent operating system It connects AI assistants to external tools and data sources through a standardized interface.

How do I install OpenFang?

Follow the installation instructions on the OpenFang GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with OpenFang?

OpenFang works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is OpenFang free to use?

Yes, OpenFang is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

OpenFang Alternatives — Similar Business Applications Servers

Looking for alternatives to OpenFang? Here are other popular business applications servers you can use with Claude, Cursor, and VS Code.

n8n

189.1k

A comprehensive MCP server that provides full control over n8n automation workflows through natural language. It offers 43 tools for managing workflows, executions, credentials, and data tables, with safety features like write-mode protection and dou

LobeHub

77.5k

🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.

Jeecgboot

46.4k

AI 低代码平台,「低代码 + 零代码」双模式驱动:低代码一键生成前后端代码,零代码 5 分钟搭建系统,AI Skills 一句话画流程、设计表单、生成整套系统。内置 AI聊天、知识库、流程编排、MCP插件等,兼容主流大模型。引领「AI 生成 → 在线配置 → 代码生成 → 手工合并->AI修改」开发模式,消除 Java 项目 80% 的重复工作,提效而不失灵活。

CowAgent

44.7k

CowAgent (chatgpt-on-wechat) 是基于大模型的超级AI助理,能主动思考和任务规划、访问操作系统和外部资源、创造和执行Skills、通过长期记忆和知识库不断成长,比OpenClaw更轻量和便捷。同时支持微信、飞书、钉钉、企微、QQ、公众号、网页等接入,可选择DeepSeek/OpenAI/Claude/Gemini/ MiniMax/Qwen/GLM/LinkAI,能处理文本、语音、图片和文件,可快速搭建个人AI助理和企业数字员工。

Minds Platform

39.2k

Platform dedicated to building an open foundation for applied Artificial Intelligence, designed for people seeking production-ready AI systems they can truly control, extend and deploy anywhere.

Astrbot

32.8k

AI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨

Browse More Business Applications MCP Servers

Explore all business applications servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "openfang": { "command": "npx", "args": ["-y", "openfang"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use OpenFang?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides