librefang/librefang

LibreFang is an open-source agent operating system written in Rust. Live demo: https://flyio.librefang.ai

What it solves

LibreFang is an Agent Operating System designed to move beyond simple chatbots. It allows users to run autonomous AI agents that operate independently 24/7 on schedules—handling tasks like lead generation, social media management, and target monitoring—without requiring constant manual prompting.

How it works

Built from scratch in Rust for performance and security, the system uses a modular kernel architecture consisting of 24 crates. It employs "Hands" (autonomous capability packages) defined by manifests and system prompts to execute specific roles. The platform integrates a wide array of LLM providers and messaging channels, utilizing a WASM sandbox and multiple security layers to ensure safe execution.

Who it’s for

It is intended for developers and power users who want to deploy persistent, autonomous agents that can interact with various third-party platforms and APIs, as well as those seeking a highly secure, open-source alternative to Python-based agent frameworks.

Highlights

  • Extensive Integrations: Supports 45 messaging channel adapters (e.g., Telegram, Discord, Slack) and 28 LLM providers.
  • Security-First Design: Features 16 security layers, including a WASM sandbox, taint tracking, and Ed25519 signing.
  • Developer Friendly: Provides an OpenAI-compatible API, client SDKs for JavaScript, Python, Rust, and Go, and a native Tauri 2.0 desktop app.
  • Interoperability: Supports the Model Context Protocol (MCP) and Google's Agent-to-Agent (A2A) protocol for cross-system delegation.
  • High Performance: Written in Rust with a comprehensive test suite (2,100+ tests) and zero clippy warnings.

Related

  • Project
    RightNow-AI/openfang

    An open-source Agent Operating System built in Rust that runs autonomous agents on schedules to perform complex background tasks without constant prompting.

  • Project
    qhkm/zeptoclaw

    A fast, secure, and local-first personal AI assistant infrastructure written in Rust, providing sandboxed autonomy and multi-provider LLM support in a single 6MB binary.

  • Project
    zeroclaw-labs/zeroclaw

    A self-hosted AI agent runtime written in Rust that connects various LLM providers to multiple communication channels and local system tools while prioritizing user data ownership.

  • Project
    vinhnx/VTCode

    An open-source Rust terminal coding agent that provides a secure, TUI-based environment for interactive and autonomous AI-driven software development.

  • Project
    liquidos-ai/AutoAgents

    A production-grade multi-agent framework in Rust that provides a type-safe environment for building intelligent systems with structured tool calling and pluggable LLM backends.