openfang
Open-source Agent Operating System
claude mcp add --transport stdio rightnow-ai-openfang openfang start \ --env PORT="4200" \ --env OPENFANG_LOG_LEVEL="info"
How to use
OpenFang is an open-source Agent Operating System written in Rust. It ships as a single binary (~32MB) that runs autonomous agents called Hands, which execute on schedules and perform complex tasks such as researching competitors, building knowledge graphs, scoring leads, monitoring targets, and generating reports. Hands are pre-built capability packages with their own manifests, system prompts, and guardrails, enabling fully autonomous workflows without manual prompting after activation. To get started, install OpenFang, initialize your environment, and start the dashboard to observe your agents in action. The system is designed to run continuously, 24/7, and you can manage Hands, view dashboards, and review results directly from the provided interface.
Once running, you can activate Hands, check their status, and configure schedules. For example, you can activate a lead-generation hand to run daily, pause a hand without losing state, and inspect all available Hands. The OpenFang ecosystem emphasizes autonomy and reliability: agents wake up, perform their tasks, and report results to your dashboard automatically. The tooling focuses on hands rather than chat prompts, making it suitable for long-running autonomous workflows such as monitoring, forecasting, and reporting.
How to install
Prerequisites:
- A modern OS (Linux, macOS, or Windows with PowerShell support)
- curl or a similar downloader
- Optional: Rust toolchain if you plan to build from source
Step-by-step install:
-
Install the OpenFang installer and run the install script:
curl -fsSL https://openfang.sh/install | sh
-
Initialize the OpenFang environment and start the platform:
openfang init openfang start
-
Access the dashboard at http://localhost:4200. If you’re on Windows, follow the Windows-specific commands shown in the documentation to install via PowerShell and then run the init/start commands.
-
(Optional) Build from source if you want to customize:
clone the repository
git clone https://github.com/RightNow-AI/openfang.git cd openfang
build the Rust binary
cargo build --release
run the built binary from target/release/openfang
./target/release/openfang start
Additional notes
Tips and common issues:
- The binary is a single ~32MB executable; ensure you have sufficient disk space and no conflicting binaries named openfang in your PATH.
- The dashboard runs on port 4200 by default; if you need to change it, set the PORT environment variable when starting the server.
- Hands are pre-built capability packages; explore Hands like Lead, Researcher, Browser, and Twitter to understand what autonomous actions OpenFang can perform.
- Windows users can use the PowerShell install path provided in the docs; ensure PowerShell is up to date for script execution.
- If you encounter instability on v0.1.0, pin to a specific commit or commit hash in production as recommended in the release notes.
Related MCP Servers
mcp-agent
Build effective agents using Model Context Protocol and simple workflow patterns
moltis
A Rust-native claw you can trust. One binary — sandboxed, secure, auditable. Voice, memory, MCP tools, and multi-channel access built-in.
flyto-core
The open-source execution engine for AI agents. 412 modules, MCP-native, triggers, queue, versioning, metering.
MakeMoneyWithAI
A list of open-source AI projects you can use to generate income easily.
pctx
pctx is the execution layer for agentic tool calls. It auto-converts agent tools and MCP servers into code that runs in secure sandboxes for token-efficient workflows.
packt-netops-ai-workshop
🔧 Build Intelligent Networks with AI