Agent Systems

Core systems for building, coordinating, and operating AI agents.

381 Projects 2 Subcategory 90 Tags
Tracked

This is a legacy category URL. Results are aggregated using the new taxonomy.

Frameworks for single-agent and multi-agent development.

A2UI

An open-source declarative UI specification and toolkit that lets agents describe renderable interfaces as safe, portable JSON.

-- Loading score

Adala

An autonomous data labeling agent framework for building adaptable data pipelines and skills.

-- Loading score

Aden Hive

A production-ready framework and runtime for building self-evolving AI agents.

-- Loading score

adk-go

A code-first Go toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.

-- Loading score

Agent Framework

Microsoft's multi-language framework for building, orchestrating, and deploying AI agents and multi-agent workflows.

-- Loading score

Agent Lightning

Agent Lightning is an open-source framework from Microsoft Research for training and improving AI agents with minimal code changes.

-- Loading score

Agent Skills

Agent Skills is an open-source collection of production-grade engineering skills by Google Chrome engineer Addy Osmani, featuring 20 structured workflows and 7 slash commands covering the full development lifecycle from spec to ship.

-- Loading score

Agent Skills

A collection that packages reusable skills as instructions and scripts to extend agents' capabilities.

-- Loading score

Agent Skills

An open format and documentation for describing, sharing, and discovering agent skills.

-- Loading score

Agent Zero

An open-source, extensible agent framework that supports multi-agent cooperation, persistent memory, and tool-enabled execution.

-- Loading score

AgentField

Brings Kubernetes principles to agent runtimes, offering an identity-aware, observable, and scalable platform for agent microservices.

-- Loading score

Agentset

An open-source platform for retrieval-augmented generation (RAG) that simplifies multi-format ingestion, partitioning, and citation-aware retrieval.

-- Loading score

Agor

Agor is a multiplayer spatial canvas from Preset for coordinating parallel AI assistant sessions and Git-linked worktrees.

-- Loading score

AionUi

A frontend UI framework and component library for LLM and agent interactions, offering customizable components, renderers, and CLI tooling for local deployment and integration.

-- Loading score

Astron Agent

An enterprise orchestration platform for building next-generation SuperAgents with agent workflows and low-code integration.

-- Loading score

Astron RPA

An agent-ready RPA suite providing out-of-the-box automation tools and enterprise integration capabilities.

-- Loading score

AWorld

AWorld is an agent runtime and research platform designed for large-scale multi-agent self-improvement and training.

-- Loading score

AWS MCP

MCP defines a standardized protocol for models, tools and hosts to securely exchange context and capability descriptions.

-- Loading score

Beads

A lightweight framework that provides persistent memory and efficient retrieval for code agents.

-- Loading score

Blades

A multimodal AI Agent framework in Go providing pluggable components for models, tools, memory, and middleware.

-- Loading score

BMAD-METHOD

BMAD-METHOD is an Agile AI-driven development method and framework that provides a toolset and best practices for building multi-role collaborative agents and engineering workflows.

-- Loading score

Browser Harness

Browser Harness is a minimal self-healing browser harness built on CDP that gives LLMs complete freedom to complete any browser task, with agents able to write missing functions at runtime.

-- Loading score

cagent

A containerized runtime and agent builder for building and running lightweight AI agents.

-- Loading score

CAMEL

CAMEL is an open-source framework for large-scale multi-agent research, supporting simulation, data generation, and collaborative agent capabilities.

-- Loading score

ChatDev

A multi-agent collaboration framework powered by large language models to turn natural-language ideas into runnable software engineering workflows.

-- Loading score

Claude Agent SDK for Python

Anthropic's Claude Agent Python SDK provides libraries and tools to interact with Claude Agent for interactive queries, tool invocation, and integrations.

-- Loading score

Claude-Flow: Orchestration Platform for Claude

Claude-Flow is an orchestration and multi-agent coordination framework for the Claude platform, supporting high-concurrency agent scheduling, Hooks automation, and Flow Nexus cloud integration.

-- Loading score

Cloudflare Agents

Cloudflare Agents: An open-source edge AI agent framework by Cloudflare that provides state management, real-time communication, and extensibility.

-- Loading score

Composio

SDKs and a platform that connect agents and LLMs to 500+ integrations, offering TypeScript and Python SDKs plus the MCP-based Rube server.

-- Loading score

CopilotKit

CopilotKit is an open-source frontend framework for rapidly integrating AI copilots, agents, and generative UI components into React and Angular applications, and the creator of the AG-UI Protocol.

-- Loading score

CUA

CUA provides an open-source infrastructure and toolchain for training and evaluating computer-use agents that operate full desktop environments (macOS, Linux, Windows).

-- Loading score

CUGA

An open-source generalist agent for enterprise, supporting web/API execution, OpenAPI/MCP integrations, composable architecture, and policy-aware features.

-- Loading score

Deep Agents

A LangChain library for building deep agents that combine planning, subagents, filesystem tools and persistent memory for multi-step reasoning.

-- Loading score

DeerFlow

DeerFlow is an open-source SuperAgent harness from ByteDance that handles different levels of tasks taking minutes to hours through the coordination of sandboxes, memories, tools, skills, and subagents.

-- Loading score

Dexter

Dexter is an autonomous agent for deep financial research that automates data collection, analysis and strategy validation.

-- Loading score

Everywhere

A context-aware desktop AI assistant that integrates multiple local and remote LLMs and MCP tools for intelligent interaction and automation.

-- Loading score

FinRobot

An open-source AI agent platform for financial analysis that integrates multi-source data, tools and large language models.

-- Loading score

Graphiti

Graphiti is an open-source framework for building real-time knowledge graphs tailored for AI agents, designed for dynamic data, agent memory, and low-latency hybrid retrieval.

-- Loading score

Hashbrown

A framework for running AI agents in the browser with Angular and React integrations.

-- Loading score

Inkeep Agents

Inkeep Agents is a framework for creating AI agents via a no-code visual builder and a TypeScript SDK, with full two-way sync for shipping assistants and multi-agent workflows.

-- Loading score

IronClaw

OpenClaw-inspired Rust implementation focused on privacy and security

-- Loading score

Lark CLI

Lark CLI is the official command-line tool for Lark/Feishu, maintained by the larksuite team. It covers 17 business domains with 200+ commands and 24 AI Agent Skills, designed for both humans and AI agents.

-- Loading score

LightAgent

An open-source, lightweight agent framework with memory, tools, and Tree-of-Thought support, designed for multi-model and multi-agent scenarios.

-- Loading score

MAI-UI

A GUI-centric agent framework supporting models ranging from 2B to 235B to build interactive agent experiences for real-world tasks.

-- Loading score

Mastra

TypeScript-based AI agent and assistant framework with workflow, tool, RAG, and observability integration.

-- Loading score

mcp-agent

A lightweight, composable agent framework built around the Model Context Protocol (MCP) to quickly assemble multi-agent, tool-enabled workflows.

-- Loading score

MiroThinker

An open-source search agent for tool-augmented reasoning, supporting very long contexts and high-frequency tool calls.

-- Loading score

Monty

A minimal, secure Python interpreter written in Rust for safely executing LLM-generated Python code.

-- Loading score

NanoClaw

NanoClaw is a lightweight containerized Claude assistant focused on secure local workflows, readable code, and fast customization.

-- Loading score

NOFX

An open-source AI trading operating system supporting multi-exchange connectivity, multi-model competition and self-evolving strategy pipelines.

-- Loading score

Open Deep Research

An open-source deep research agent framework that integrates multi-model providers, search tools and MCP for reproducible research pipelines.

-- Loading score

OpenAgents

OpenAgents is an open-source platform for creating and connecting AI agent networks, supporting multiple protocols and plugin extensions.

-- Loading score

OpenClaw

A local-first personal AI assistant that runs on your devices and integrates with messaging channels.

-- Loading score

OpenHands

An open-source developer platform that uses autonomous agents to assist with code editing, command execution and testing.

-- Loading score

OpenManus

OpenManus is a modular open-source agent framework that helps move natural-language driven agent prototypes into deployable engineering systems.

-- Loading score

OxyGent

A multi-agent collaboration framework for enterprise applications that emphasizes local-first workflow composition and tool integration.

-- Loading score

Pipelex

An open-source language and toolkit for building and running reproducible AI agent workflows.

-- Loading score

Planning with Files

A Manus-inspired workflow that uses persistent Markdown files to manage plans and agent skills.

-- Loading score

Qwen-Agent

Qwen-Agent is an open-source agent framework that provides tool calling, RAG, code interpreter and deployment examples to quickly build intelligent assistants and applications.

-- Loading score

Ralph for Claude

An open-source autonomous development loop toolkit for Claude Code, providing session continuity, rate limiting and circuit breaker protections.

-- Loading score

ruflo

ruflo is the leading agent orchestration platform for Claude, enabling deployment of intelligent multi-agent swarms, autonomous workflow coordination, and conversational AI systems with enterprise-grade architecture.

-- Loading score

Skills

An open-source library from Anthropic for defining, sharing, and reusing task-oriented capability modules.

-- Loading score

Skyvern

Skyvern is an open-source platform that combines vision and LLMs to automate browser-level workflows, available as local software and a managed cloud service.

-- Loading score

Stakpak Agent

A terminal-native DevOps agent written in Rust that can run commands, edit files, search docs, and generate high-quality IaC.

-- Loading score

Stripe AI

An open-source collection of SDKs and tools from Stripe that help integrate payments and billing into LLMs and agent frameworks.

-- Loading score

Superpowers

An open-source development workflow and skills library for coding agents that emphasizes TDD, process, and verifiable automation.

-- Loading score

Swarms

A production-ready multi-agent orchestration framework that provides scalable runtimes and protocols for collaborative agents.

-- Loading score

SWE-agent

SWE-agent: an academic-developed automated software engineering agent framework for code repair, benchmarking, and research workflows.

-- Loading score

TradingAgents

TradingAgents is a multi-agent framework for financial trading that leverages LLM-driven strategies and backtesting tools.

-- Loading score

tRPC-Agent-Go

tRPC-Agent-Go is a production-ready Go framework for building intelligent agent systems with multi-agent orchestration, tooling, and observability.

-- Loading score

UFO: Desktop AgentOS

UFO is an open-source Desktop AgentOS for Windows that turns natural-language requests into reliable, multi-application workflows.

-- Loading score

Universal Commerce Protocol (UCP)

A standardized protocol that enables AI agents, platforms, and commerce providers to interoperate securely and consistently.

-- Loading score

Vibe Skills

Vibe Skills is an open-source all-in-one AI skill package that integrates expert-level capabilities and context management, enabling any AI agent to instantly upgrade its functionality.

-- Loading score

VM0

A runtime and orchestration platform for natural-language-described agents, offering session persistence, observability, and multi-model routing.

-- Loading score

xmcp

A TypeScript framework for building and shipping MCP servers that streamlines developer experience and lowers the barrier to create tools on the Model Context Protocol ecosystem.

-- Loading score

Youtu-Agent

Youtu-Agent is an open-source agent framework published by Tencent Cloud, aimed at research and engineering use.

-- Loading score

Task routing, workflow orchestration, and execution control.

AutoGen

A programming framework for agentic AI that enables development of multi-agent workflows with a layered and extensible design.

-- Loading score

BettaFish (WeiYu)

An open-source multi-agent platform for automated collection, analysis, and reporting of massive social media data.

-- Loading score

Claude Code Agents & Plugins

Claude Code Agents is an ecosystem of agents and plugins for the Anthropic Claude Code platform, enabling multi-agent orchestration, automation, and efficient collaboration.

-- Loading score

Codebuff

Discover Codebuff, the open-source AI coding assistant that enhances project management with multi-agent architecture for safer, automated code modifications.

-- Loading score

CrewAI

Discover CrewAI, the open-source Python framework for high-performance multi-agent orchestration, enabling seamless collaboration and customizable workflows.

-- Loading score

DeepCode

An open-source multi-agent platform that converts papers and natural language into production-ready code implementations.

-- Loading score

Eigent

An open-source desktop multi-agent platform that supports local deployment, MCP tool integrations, and enterprise features to automate complex workflows and boost productivity.

-- Loading score

ElizaOS

ElizaOS is an open-source, extensible platform for building, deploying, and managing multi-agent AI applications, offering CLI tooling, a web dashboard, and plugin-based extensibility.

-- Loading score

GPT Researcher

A deep-research multi-agent framework that automates web and local document retrieval to produce sourced research reports.

-- Loading score

HiClaw

Open-source Agent Teams system with IM-based multi-Agent collaboration and human-in-the-loop oversight

-- Loading score

JoyAgent-JDGenie

JoyAgent-JDGenie is an open-source end-to-end multi-agent framework for task orchestration and report generation.

-- Loading score

MetaGPT

Multi-agent framework that automates software development processes from requirements to code implementation through collaborative AI agent teams.

-- Loading score

OpenAI Agents (Python)

OpenAI's lightweight Agents SDK for Python—build multi-agent workflows with agents, handoffs, guardrails, tracing, and session management. Provider-agnostic and extensible for production use.

-- Loading score

OpenMAIC

OpenMAIC is an open-source AI platform that transforms any topic or document into a rich, interactive classroom experience through multi-agent orchestration. It features one-click lesson generation, AI teachers and classmates for real-time interaction, rich scene types (slides, quizzes, interactive simulations, and project-based learning), whiteboard drawing, voice explanations, and export to PPTX or HTML formats.

-- Loading score

OWL

OWL (Optimized Workforce Learning) is an open-source framework for multi-agent collaboration and task automation, supporting tool invocation, browser automation, and multimodal processing.

-- Loading score

Semantic Kernel

A model-agnostic SDK for building, orchestrating, and deploying scalable AI agents and multi-agent systems.

-- Loading score

Sympozium

The coordination layer for multi-agent AI, providing orchestration and communication infrastructure for agent teams.

-- Loading score

ValueCell

A community-driven multi-agent platform for finance that offers research, strategy development, and automated trading while keeping sensitive data local.

-- Loading score