Stay up-to-date with the latest in AI, AGI, and AI-agent projects by subscribing to our newsletter: Subscribe here
This file is updated weekly with the newest AI-agent projects from the open-source community.
An AI-powered second brain to help you organize and retrieve knowledge across sources like tweets, bookmarks, articles, and websites—built with memory-first AI agents for personal productivity.
GitHub Link
A powerful collection of templates and examples to build generative AI agents using LangChain, LangGraph, and LLM frameworks. Helps you deploy intelligent agents that can reason, retrieve, and take actions.
GitHub Link
A modular AI Agent platform for financial applications. Supports intelligent workflows for stock analysis, financial QA, and more—powered by large language models and built for scalability.
GitHub Link
Composable agent framework tailored for Model Context Protocol (MCP). Enables multi-agent orchestration, modular tool integration, and reusable reasoning templates for LLM-powered workflows.
GitHub Link
A self-evolving meta-agent that can build and refine other AI agents. Uses internal tool libraries and memory mechanisms, powered by MCP and designed for agent self-improvement.
GitHub Link
A lightweight browser-based interface to interact with AI agents in real time. Built with Gradio and LangGraph, it supports persistent sessions and multiple LLM backends.
GitHub Link
Connects LLM agents to browsers via Playwright automation and the MCP protocol. Allows real-time website interaction, test automation, and dynamic page navigation using AI.
GitHub Link
Microsoft’s approach to fusing knowledge bases into LLMs without retrieval. Uses “knowledge tokens” for seamless in-context learning, ideal for enterprise and agentic tasks.
GitHub Link
A Go language implementation of the Model Context Protocol. Build MCP-compatible tools, agents, and systems for LLM workflows using fast, statically typed development.
GitHub Link
FastMCP offers a high-level Pythonic implementation of MCP. Ideal for quickly connecting tools, prompts, and resources to your LLM agents in structured, scalable workflows.
GitHub Link
Giselle is an open-source AI designed for agentic workflows, enabling seamless human-AI collaboration to automate complex tasks and streamline operations.
GitHub Link
A robust SDK for building sophisticated multi-agent systems using OpenAI’s APIs. It enables complex interactions and seamless handoffs between specialized agents, making it a powerful tool for developing intelligent, collaborative applications.
GitHub Link
A comprehensive platform designed to build, evaluate, and monitor AI agents. Agentops focuses on observability and streamlined workflows, providing deep insights into agent performance and facilitating rapid prototyping and deployment.
GitHub Link
Developed by Microsoft, RDAgent is an intelligent framework aimed at automating remote desktop or data tasks through AI-driven processes. It helps streamline complex operations, reducing manual intervention with smart automation.
GitHub Link
LARS enhances AI agent reasoning and task management. This flexible system enables agents to perform sophisticated operations with minimal supervision, making it a valuable asset for building autonomous workflows.
GitHub Link
MetaGPT is an advanced multi-agent framework that simulates an entire AI software company. It orchestrates diverse agent roles—from product management to engineering—allowing agents to collaborate on comprehensive projects with end-to-end autonomy.
GitHub Link
Mem0 provides a dedicated memory layer for AI agents, enabling them to retain context across interactions. This persistent memory capability enhances personalization and effectiveness, ensuring more coherent and relevant agent responses.
GitHub Link
OpenHands is an open platform that empowers AI agents to autonomously modify code, execute commands, and integrate with various services. It streamlines development processes and supports a broad range of intelligent automation tasks.
GitHub Link
Agno is a lightning-fast, model-agnostic framework for building multimodal AI agents. It emphasizes speed and efficiency, offering rapid agent creation with minimal memory overhead and seamless integration with any language model provider.
GitHub Link
Executive AI Assistant automates executive tasks with intelligent, autonomous workflows, managing emails, scheduling, and more with minimal human intervention.
GitHub Link
AstrBot is an AI agent framework that facilitates intelligent collaboration between agents to automate complex tasks, streamlining workflows in multi-agent environments.
GitHub Link
Browser-use enables AI agents to interact directly with web browsers, automating online tasks and bridging the gap between AI and web interfaces for seamless automation.
GitHub Link
Page Assist integrates AI into your browsing experience by allowing agents to extract, analyze, and interact with web content locally, enhancing online productivity.
GitHub Link
The AI Scientist is a groundbreaking system that empowers autonomous scientific discovery by enabling AI agents to generate research ideas, conduct experiments, and even perform automated peer review.
GitHub Link
LangManus is an open-source automation framework that equips AI agents with a robust toolkit—combining LLMs with specialized tools—to orchestrate complex, multi-agent workflows efficiently.
GitHub Link
Local Deep Researcher provides a fully local web research assistant that leverages locally hosted LLMs to perform in-depth, privacy-preserving research and generate detailed summaries.
GitHub Link
Sidekick is a macOS application that delivers a local, privacy-focused AI agent, enabling offline interactions and intelligent local data retrieval without relying on cloud services.
GitHub Link
LeRobot democratizes AI for robotics by offering tools that enable autonomous control and decision-making in real-world robotic applications, making advanced robotics accessible to all.
GitHub Link
Designed as an entry point into the world of autonomous AI, this project offers tutorials and sample projects to help you build, deploy, and understand AI agents from scratch.
GitHub Link
AutoAgent streamlines the development and management of AI agents by automating the creation of dependency graphs and executable workflows, making agent deployment simpler than ever.
GitHub Link
CSM (Conversational State Machine) provides a framework for building structured conversational agents with efficient state management and dynamic interaction capabilities.
GitHub Link
AutoMate is an automation tool that leverages AI agents to execute routine tasks across various workflows, offering a hands-free approach to daily operations.
GitHub Link
Agent Chat UI offers an intuitive interface for interacting with multiple AI agents simultaneously, enabling smooth communication, collaboration, and task management.
GitHub Link
Realtime API Agents is an innovative project built on OpenAI’s Realtime API. It demonstrates advanced multi-agent patterns for interactive voice and conversational applications by orchestrating several specialized agents. With dynamic handoffs and real‐time decision making, this project paves the way for building robust, autonomous voice applications.
GitHub Link
CopilotKit is a framework that enables developers to build AI-powered coding assistants. By integrating with popular development environments, it helps automate code generation, debugging, and project management tasks. Its intuitive design bridges the gap between human expertise and machine efficiency.
GitHub Link
Codename Goose is an autonomous AI engineering agent that automates complex development tasks. It intelligently handles debugging, dependency management, and routine code generation to boost productivity and streamline your engineering workflow—all while learning from your project context.
GitHub Link
Integuru is an AI agent that builds third-party integrations by reverse-engineering internal APIs. By capturing network requests and generating executable dependency graphs, Integuru automates tasks like document retrieval and data extraction, dramatically reducing manual integration efforts.
GitHub Link
TheAgentCompany offers a benchmark framework for evaluating AI agents on real-world professional tasks. It simulates a digital workplace where agents perform diverse roles—providing insights into their strengths and weaknesses in authentic enterprise scenarios.
GitHub Link
VisionAgent is an AI-powered tool designed to generate code for computer vision tasks. By translating high-level instructions into functional code for object detection, tracking, and image analysis, VisionAgent simplifies the development of vision applications using cutting-edge LLMs.
GitHub Link
Prompt-To-Agent bridges the gap between human language and autonomous agent actions. It converts natural language prompts into executable behaviors, allowing developers to create custom AI agents without deep coding expertise—accelerating the prototyping and deployment of intelligent workflows.
GitHub Link
TEN Agent is a lightweight framework that enables you to build and deploy autonomous AI agents with ease. It offers an intuitive API for creating agents that can perform complex tasks seamlessly.
GitHub Link
Agenta provides a robust platform for developing AI agents that can integrate with various applications. It streamlines the process of building intelligent, autonomous agents with a focus on ease-of-use and scalability.
GitHub Link
Steel is an open-source browser API designed specifically for AI agents and applications. It simplifies web automation and interaction, allowing agents to operate within a secure, containerized browser environment.
GitHub Link
PR-Agent is built to automate pull request processes using AI, helping streamline code reviews and integration workflows. It leverages natural language processing to understand and act on repository changes.
GitHub Link
Farcaster Agent empowers developers to create AI-powered agents that interact with social platforms like Farcaster. It facilitates seamless communication and automation across decentralized networks.
GitHub Link
Terminal GPT brings the power of large language models directly into your command line, enabling interactive AI conversations and task automation right from the terminal.
GitHub Link
Qwen-Agent offers a robust framework for building agents powered by the Qwen language models. It focuses on efficient task management and responsive, context-aware interactions.
GitHub Link
Open Interface transforms your computer into a self-driving system by allowing AI agents to control keyboard and mouse actions via natural language commands, bridging human intent with computer execution.
GitHub Link
ZerePy serves as a launchpad for creating and deploying AI agents that can engage on social platforms and manage blockchain interactions. Its modular design and comprehensive integration options simplify agent development.
GitHub Link
AppAgent empowers developers to create multimodal AI agents that interact directly with smartphone applications. It delivers context-aware assistance and automates routine tasks on mobile devices.
GitHub Link
HuatuoGPT-o1 is revolutionizing medical reasoning by enabling AI agents to assist with healthcare diagnostics and treatment recommendations. Its intelligent analysis and feedback capabilities make it a promising tool for medical applications.
GitHub Link
Agent.exe empowers AI to control your computer by integrating with powerful language models. It offers seamless automation and intuitive command execution—giving you a hands-free experience while still keeping full control at your fingertips.
GitHub Link
GPT PILOT leverages advanced GPT models to serve as an intelligent assistant, automating complex tasks and streamlining workflows. This project helps you harness the power of large language models for practical, everyday applications with minimal manual input.
GitHub Link
This template provides a robust foundation for building retrieval-based AI agents using LangChain. It enables developers to integrate efficient information retrieval within conversational agents, making it easier to create smart, context-aware systems.
GitHub Link
Chat Nio is a next-generation AI chat platform that integrates multiple AI models into a unified interface. It delivers intelligent, context-aware interactions, making conversational experiences smoother and more engaging than ever before.
GitHub Link
Formerly known as MemGPT, Letta is an agent framework that empowers developers to build stateful AI applications with memory. By maintaining context over interactions, Letta enables more natural and coherent conversations, opening up new possibilities for personalized AI assistants.
GitHub Link
Multi-Agent Orchestrator is a framework designed to coordinate multiple AI agents. It intelligently routes tasks among specialized agents, enabling dynamic collaboration and creating richer, more responsive conversational systems.
GitHub Link
Luna Executor is a powerful tool for Roblox that automates the execution of custom scripts. It enhances gameplay by streamlining complex in-game tasks and offers a user-friendly interface for seamless code deployment—truly unleashing your scripting power.
GitHub Link
DuckDuckGPT integrates advanced GPT-powered responses directly into DuckDuckGo searches. It enriches your search experience with detailed, AI-generated answers while keeping your privacy intact, making it a unique blend of search and conversational AI.
GitHub Link
Agent Zero is an open-source framework designed for building autonomous AI agents that perform a range of tasks with minimal human intervention.
GitHub Link
TinyAgent is a lightweight framework that brings AI capabilities to edge devices, enabling on-device processing and function calling for autonomous operations.
GitHub Link
AutoGPT is an experimental open-source application that showcases the power of autonomous AI agents by chaining together large language model actions to accomplish complex goals.
GitHub Link
AppAgent empowers developers to create AI agents that interact with smartphone applications, delivering context-aware assistance and automating routine mobile tasks.
GitHub Link
GenAI Agents streamlines the process of developing and deploying generative AI agents, leveraging large language models to build intelligent, responsive systems.
GitHub Link
LiveKit Agents is a real-time framework for building multimodal AI applications, allowing agents to process audio and video streams and integrate seamlessly with live communication systems.
GitHub Link
For previous updates, please refer to our commit history or check the changelog.