AI agent orchestration / agent OS
AGNT is a local-first agent operating system that provides a desktop app and self-hostable backend for building, running, and improving AI agents, workflows, goals, memory, and evals.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
Choose it as a self-hosted TypeScript runtime if you want a drop-in multi-agent framework with broader LLM provider support.
Open-source self-hosted AI agent runtime and multi-agent framework for autonomous agent swarms. Agent memory, MCP tools, schedules, delegation, and 23+ LLM providers (Claude, GPT, Gemini, OpenRouter, Ollama). A practical Claude Code and LangChain alternative.
Choose it for a TypeScript agent harness with swarm orchestration if you want a code-first alternative to the desktop app.
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
Choose it for a Rust agent harness with local-first memory if you need a systems-language alternative to the seed's JavaScript stack.
Your Personal AI super intelligence. A brain that builds a local-first memory of your life, a fantastic orchestrator of agent fleets and workflows, and a deep researcher.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
Choose it if you prefer a Python GPL-3.0 extensible AGI framework over a JavaScript local-first OS.
Extensible AGI Framework
Choose it for a Python self-improving agent library if you want a simpler runtime than AGNT's full workspace.
self-improving agent
Choose it for a Python agent harness with a large ecosystem if you don't need AGNT's desktop app.
The agent that grows with you
Choose it for a cloud-hosted autonomous agent platform if you want a managed service instead of self-hosting.
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
Choose it for a production-grade multi-agent harness in Python if you need enterprise deployment rather than local-first desktop.
Multi-Agent Harness for Production AI
Choose it to run unattended agents on GitHub Actions if you prefer a shell-based framework over a desktop app.
The most autonomous AI agent framework: runs unattended on GitHub Actions, self-healing skills, drives Claude Code, Grok, Codex & more. No approval loops. Configure once, forget forever.
Choose it for Python research on vision-grounded memory and self-written skills if you need those specific capabilities.
Autonomous self-evolving agents. Vision-grounded layered memory and self-written skills for LLM agents that operate your computer.
Choose it for a TypeScript platform-engineering focus with governance features if that's your domain.
CAIPE is an open-source AI platform for building, governing, and operating AI agents and agentic workflows for platform engineering and beyond.
Choose it for a Python/TypeScript SDK to build your own harness if you want library-level control instead of a full OS.
Build an agent harness and control it end-to-end. Open-source SDK for production AI agents in Python & TypeScript - any model, any cloud.
Choose it for a .NET library to build agents if your stack is C# rather than JavaScript.
The .NET library to build AI agents with 30+ built-in connectors.
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Choose it for a lean TypeScript coding agent when you don't need AGNT's visual workflow builder or goal system.
A self-improving RLM agent for coding workflows and long-running autonomous tasks.
Choose it when you need a mature software-engineering agent, accepting a narrower scope than AGNT's general workflows.
🙌 OpenHands: AI-Driven Development
Choose it for deterministic AI coding harnesses if your work is code-centric rather than general agent workflows.
The first open-source harness builder for AI coding. Make AI coding deterministic and repeatable.
Choose it for a Go-based large-project coding agent, accepting a slower release cadence than the seed.
Open source AI coding agent. Designed for large projects and real world tasks.
Choose it for a lightweight Rust implementation of the Ralph loop if you only need autonomous orchestration, not a full OS.
An improved implementation of the Ralph Wiggum technique for autonomous AI agent orchestration
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
Choose it as a ready-to-use personal assistant app when you don't need to build or self-host an agent OS.
⚡️next-generation personal AI assistant powered by LLM, RAG and agent loops, supporting computer-use, browser-use and coding agent, demo: https://demo.openagentai.org
Choose it as a terminal-based personal agent for ad-hoc local tasks instead of a full orchestration OS.
Your agent in your terminal, equipped with local tools: writes code, uses the terminal, browses the web. Make your own persistent autonomous agent on top!
Choose it as a ready-to-use self-programming assistant for workflow capture and automation, not a buildable OS.
Self-Programming AI Assistant. Capture, automate, and refine all your workflows.
Choose it as a local-first desktop app if you want a GUI orchestrator without building from source.
Open-source multi-agent AI desktop client — build and command your AI agent team through conversation. A commander LLM dispatches sub-agents in parallel or in series; agents self-evolve via reflection and skill crystallization. Local-first, BYO LLM keys (Claude · OpenAI · Gemini · DeepSeek · Kimi · GLM · Qwen). macOS / Windows / Linux.
Choose it as a local-first workspace app for multi-agent chat and visual workflows if you prefer a packaged GUI.
Web dashboard for Hermes Agent — multi-platform AI chat, session management, scheduled jobs, usage analytics
These projects were analysed and named agnt among their alternatives. The relationship is not symmetric — how agnt rates them is a separate judgement, made when agnt is analysed in its own right.
calls agnt “choose it for a local-first OS to build and run AI agents and workflows, though it is broader and not coding-specific”
The first open-source harness builder for AI coding. Make AI coding deterministic and repeatable.
calls agnt “Choose it if you want a local-first runtime for many agent types and workflows rather than a coding-specific RLM REPL harness.”
A self-improving RLM agent for coding workflows and long-running autonomous tasks.
calls agnt “choose for a local-first agent OS with workflows and scheduled goals instead of a single prompt loop”
Autonomous AI development loop for Claude Code with intelligent exit detection
calls agnt “Choose for a local-first JavaScript agent runtime if you are not tied to Python.”
BabyAGI: an Autonomous and Self-Improving agent, or BASI
calls agnt “JavaScript local-first agent OS; pick it to build and debug agents on a workstation rather than in scheduled cloud jobs.”
The most autonomous AI agent framework: runs unattended on GitHub Actions, self-healing skills, drives Claude Code, Grok, Codex & more. No approval loops. Configure once, forget forever.