RightNow-AI/openfang
OpenFang is an open-source agent operating system, built in Rust, designed to facilitate the management and orchestration of AI agents.
Awesome AI Agents › AI Agent Operating System
compozy/compozy
CompozyOS is a local-first runtime that gives agent work a durable place to live. A single home-scoped daemon holds sessions, tasks, Loop runs, memory, permissions, automation and tools, so closing one client does not erase the work in progress. People steer that state through a web shell and a CLI, while agents operate the same resources through structured HTTP and server-sent events, a Unix domain socket, MCP and native tool surfaces. Because every surface reads daemon-owned state, clients do not maintain a parallel model of what is happening. The daemon resolves the workspace, applies permissions and runtime policy, coordinates ACP agents and persists events and resource state; approvals, permissions, run state, artifacts and agent activity stay inspectable while work continues in the background. The project ships as one Go binary with SQLite-backed stores, keeping runtime state on the local machine unless a configured provider or extension owns an external boundary. A separate Gateway component pairs devices and exposes only the private or public surfaces that are explicitly enabled. Commands such as daemon start, status and daemon stop manage the local process, and structured JSON output is available for programmatic callers. Task Schema v2 keeps authored task files as portable Markdown with typed frontmatter; the runtime imports them into durable tasks and executes them through Loops. Configuration layers global defaults in a home config file under workspace overrides and explicit command flags. Reusable agent definitions live in per-name directories with an AGENT.md and an optional agent-local mcp.json, and workspace definitions override global ones as a whole. Extensions add versioned resources and runtime behaviour through declared provide surfaces, with the daemon owning discovery, enablement, trust decisions, lifecycle and hooks; version-matched SDKs are published for Node and Go. Sessions can also discover peers, exchange typed messages and delegate work over the compozy-network protocol. The v0.3 line is in beta.
https://github.com/compozy/compozy
Also in AI Agent Operating System
Arindam200/awesome-ai-apps
A comprehensive repository showcasing practical examples, tutorials, and tools for building powerful AI applications using large language models and various AI agent frameworks.
OpenByteInc/QuantDinger
QuantDinger is an open-source AI trading operating system designed for independent traders, Python strategy authors, and small teams, offering backtesting, live trading, and multi-agent research ca...
MemTensor/MemOS
MemOS is an operating system for Large Language Models that enhances their capabilities with long-term memory, enabling more context-aware, consistent, and personalized AI interactions through a modular and extensible memory architecture.
holaboss-ai/holaOS
HolaOS is a local-first, AI-powered desktop agent that learns your work context, integrates with over 100 workplace tools, and manages memory efficiently to provide proactive assistance without for...
mcp-use/mcp-use
mcp-use is an open-source unified client library that enables seamless integration of any LangChain-supported LLM with MCP servers to build custom agents with tool access.
astrid-runtime/astrid
Astrid is a portable, capability-secure operating system for composable software components, providing an OS-grade boundary for AI agents and other untrusted code via WebAssembly capsules.
HKUDS/AutoAgent
AutoAgent is a fully-automated, zero-code framework for creating and deploying LLM agents using natural language, featuring top GAIA benchmark performance and broad LLM support.