Discover and compare all mainstream AI Agent frameworks. Real-time GitHub stats, deep reviews, and one-click comparison to help you find the perfect tool.
Sorted by GitHub Stars
Core Framework
OpenClaw is the original and most mature open-source AI Agent framework, created by Peter Steinberger. It enables 24/7 autonomous task execution, connects to chat platforms (WeChat, DingTalk, Feishu, Discord, etc.), supports multiple AI models (Claude, GPT, Kimi, GLM, etc.), and stores data locally for privacy control. With 364k+ GitHub stars, it has the largest community and skill ecosystem (ClawHub with 13,729+ skills).
Coding Assistant
An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.
Agent Platform
n8n is a fair-code workflow automation platform with native AI agent capabilities, founded in Germany. With 187K+ GitHub stars, n8n combines visual workflow building with custom code flexibility, supporting 400+ integrations. Unlike traditional automation tools, n8n natively supports AI agent creation using LangChain, enabling users to build complex AI agent workflows with visual orchestration. n8n can be self-hosted or run in the cloud, giving teams full control over their data. The platform supports MCP (Model Context Protocol), making it a key tool in the AI agent ecosystem. n8n raised funding in 2025 and is rapidly growing as an AI agent orchestration platform.
Personal Assistant
AutoGPT is the pioneering autonomous AI agent created by Significant Gravitas. As the first project to demonstrate true autonomous task planning and execution, it sparked the AI Agent revolution in 2023. With 184k+ GitHub stars, AutoGPT allows users to build and deploy AI agents that can think, plan, and act independently to achieve complex goals.
Coding Agent
OpenCode is a 100% open-source AI coding agent with 158k+ GitHub stars, used by over 1.5 million developers monthly. It works in terminal (TUI), desktop app, and IDE. Supports 75+ model providers including Claude, GPT, Gemini, DeepSeek, and local models via LM Studio — no vendor lock-in. Features client/server architecture for remote control (even from mobile), built-in LSP support with auto-loading for multiple languages (Rust, Swift, Terraform, TypeScript, PyRight), multi-session parallel execution, shareable session links for debugging, MCP server support (local and remote), privacy-first design (no code or context data stored), and Apache 2.0 license. Compatible with JetBrains IDEs, Zed, Neovim, Emacs, and more via Agent Client Protocol (ACP).
Core Framework
Hermes Agent by Nous Research is a self-improving AI Agent framework built in Python. It features a built-in learning loop with a 5-stage skill extraction mechanism, cross-session persistent user modeling, and native multi-agent support. With 95.6k+ GitHub stars, it is the fastest-growing Agent framework of 2026, offering seamless data migration with OpenClaw.
Core Framework
The most mature AIエージェントframework with the largestエコシステム
Lightweight Variant
セキュリティfirst OpenClaw variant with container isolation
Lightweight Variant
Rustベースultra軽量AIエージェントwith sandbox isolation MB RAM
Lightweight Variant
Goベースsingle binary AIエージェントfor IoT and embeddedデバイスMB RAM
Lightweight Variant
WASM sandboxセキュアエージェントOS with zero trust architecture
Lightweight Variant
Ultra軽量AIエージェント(~4000 lines of Python a minimalist alternative to OpenClaw
Your comprehensive hub for AI agent frameworks
Real-time rankings based on GitHub stars, forks, and community activity.
Detailed feature breakdowns, pros/cons analysis, and use case recommendations.
Side-by-side comparison to help you find the perfect framework.
Code generation, autocomplete, and review tools
Web interaction, scraping, and UI testing
Multi-agent collaboration and task orchestration
Enterprise-grade AI agent deployment solutions
Community-driven open-source agent projects
On-premise, data-private AI agent solutions