Skip to content
BranchesTags

Latest commit

 

History

102 Commits

Folders and files

NameName
Last commit message
Last commit date
.github
.github
 
 
data
data
 
 
scripts
scripts
 
 
.gitignore
.gitignore
 
 
AGENTS.md
AGENTS.md
 
 
CONTRIBUTING.md
CONTRIBUTING.md
 
 
LICENSE
LICENSE
 
 
README.md
README.md
 
 

Repository files navigation

Awesome Agent Evolution Awesome

AI Agent self-evolution, memory systems, autonomous self-improvement, and the infrastructure that powers them.

Contents

Taxonomy

graph LR
    Root["Agent Evolution"] --> SingleAgent["Single-Agent<br/>Optimization"]
    Root --> Infrastructure["Infrastructure<br/>& Protocols"]

    SingleAgent --> Evolution["Self-Evolution<br/>& Self-Improvement"]
    SingleAgent --> Memory["Memory<br/>Systems"]
    SingleAgent --> PromptOpt["Prompt & Behaviour<br/>Optimization"]

    Infrastructure --> Protocols["A2A & MCP<br/>Protocols"]
    Infrastructure --> Platforms["Agent Development<br/>Platforms"]
    Infrastructure --> Coding["Agent Coding &<br/>Software Engineering"]
    Infrastructure --> Safety["Agent Safety<br/>& Guardrails"]
    Infrastructure --> Embodied["Embodied AI<br/>& Device Control"]
Loading

Agent Evolution and Self-Improvement

Projects focused on enabling AI agents to evolve, learn, and improve autonomously.

  • Eliza - Autonomous agents for everyone. A framework for creating and deploying AI agents that evolve over time. by @elizaOS (19,301 stars)
  • Agent Zero - General-purpose AI agent framework that learns and evolves through interaction. by @agent0ai (19,128 stars)
  • SuperAGI - A dev-first open source autonomous AI agent framework. Build, manage and run self-improving autonomous agents. by @TransformerOptimus (17,678 stars)
  • evolver - The GEP-powered self-evolution engine for AI agents. Genome Evolution Protocol enables agents to evolve autonomously via mutation and selection. by @EvoMap (9,074 stars)
  • OpenEvolve - Open-source evolutionary coding agent inspired by AlphaEvolve. Evolves code solutions through LLM-driven mutation and selection. by @algorithmicsuperintelligence (7,339 stars)
  • Agents (aiwaves) - An open-source framework for data-centric, self-evolving autonomous language agents. by @aiwaves-cn (5,960 stars)
  • EvoAgentX - Automated framework for evolving agentic workflows. Optimizes agent prompts, tools, and pipelines via evolutionary algorithms. by @ANative-Lab (3,316 stars)
  • HyperAgents - Self-referential self-improving agents by Meta. DGM-Hyperagents add an optimization layer so agents edit their own improvement process. by @facebookresearch (2,719 stars)
  • SIA - Self-improving AI framework that autonomously optimizes the performance of any AI system through iterative evaluation and refinement. by @hexo-ai (2,146 stars)
  • Orkas - Local-first multi-agent desktop application whose specialist agents improve through reflection and crystallize successful approaches into private reusable skills. by @Orkas-AI (1,834 stars)
  • Ouroboros - Self-creating AI agent that writes its own code and evolves autonomously. Completed 30+ evolution cycles in first 24 hours with zero human intervention. by @razzant (1,283 stars)
  • Agent0 - Self-evolving agent framework from UNC/Salesforce/Stanford. Improves without human-curated datasets via curriculum and executor agent competition. by @aiming-lab (1,258 stars)
  • agent-qa - Self-improving QA agent for natural-language web and mobile tests, using persistent run memory to adapt to UI changes and catch regressions. Source-available under FSL-1.1-ALv2: use is permitted except to provide a competing commercial product or service, and each release converts to Apache-2.0 after two years. by @vostride (904 stars)
  • Reef - Continual learning infra for self-improving agents. Serves agent traffic, turns matched feedback into model-weight or harness updates, and publishes accepted updates as versioned artifacts. by @Human-Agent-Society (796 stars)
  • A-Evolve - The PyTorch for Agentic AI. Open-source infrastructure that evolves any agent across any domain with zero human intervention. #1 on MCP-Atlas (79.4%). by @A-EVO-Lab (776 stars)
  • OpenProgram - Self-programming AI agent framework whose agents create, run, and refine their own workflows while the runtime manages models, tools, memory, context, and multi-agent collaboration. by @Fzkuji (422 stars)
  • SEAgent - Self-Evolving Computer Use Agent with Autonomous Learning from Experience. by @SunzeY (262 stars)

Memory Systems

Vector, graph, episodic, and hybrid memory architectures for persistent agent cognition.

  • Mem0 - Production-ready AI agent memory with scalable long-term memory. 26% improvement over baseline on LOCOMO benchmark with 91% latency reduction. by @mem0ai (64,952 stars)
  • Cognee - Knowledge engine for AI agent memory. Build and query knowledge graphs from unstructured data in 6 lines of code. by @topoteretes (30,605 stars)
  • agentmemory - Persistent, benchmark-tuned memory for coding agents (Claude Code, Cursor, Copilot CLI, Codex, and any MCP client). Remembers context across sessions so you stop re-explaining. by @rohitg00 (28,189 stars)
  • TencentDB Agent Memory - Fully local long-term memory for AI agents via a four-tier progressive storage architecture, from Tencent Cloud. by @TencentCloud (26,200 stars)
  • Letta - Platform for building stateful agents with advanced self-editing memory. Formerly MemGPT. by @letta-ai (24,668 stars)
  • Memvid - Single-file memory layer for AI Agents in Rust. +35% SOTA on LoCoMo with ultra-low latency (0.025ms P50). by @memvid (16,528 stars)
  • memU - Memory system for 24/7 proactive agents. Persistent memory across sessions and platforms. by @NevaMind-AI (14,395 stars)
  • EverMemOS - Long-term memory for 24/7 AI agents across LLMs and platforms. by @EverMind-AI (12,803 stars)
  • holaOS - Agent environment for long-horizon work, continuity, and self-evolution. by @holaboss-ai (11,189 stars)
  • ChatLab - Rediscover your social memories with local, AI-powered analysis. 本地化的聊天记录分析工具，通过 AI Agent 回顾你的社交记忆。. by @ChatLab (7,331 stars)
  • honcho - Memory library for building stateful agents with user context management. by @plastic-labs (7,074 stars)
  • memgraph - High-performance open-source in-memory graph database for GraphRAG, AI memory, agentic AI, and real-time graph analytics. Cypher-compatible, built in C++. by @memgraph (4,455 stars)
  • Acontext - Open-source skill memory layer for AI agents. Automatically captures learnings from agent runs and stores them as reusable skill files. by @memodb-io (3,687 stars)
  • ReMe - Memory management kit for agents. File-based and vector-based memory systems. SOTA on LoCoMo and HaluMem benchmarks. by @agentscope-ai (3,435 stars)
  • MemMachine - Universal memory layer for AI agents. Episodic (graph-based), profile (SQL), and working memory with scalable storage and retrieval. by @MemMachine (3,217 stars)
  • datachain - Operational data context layer for AI agents providing typed and versioned datasets over multimodal content. by @datachain-ai (2,819 stars)
  • nocturne_memory - Lightweight, rollbackable Long-Term Memory Server for MCP Agents with graph-like structured memory. by @Dataojitori (1,346 stars)
  • Mem9 - Unlimited persistent memory layer for AI agents. Cloud-synced memory across sessions and tools. by @mem9-ai (1,208 stars)
  • Awesome-AI-Memory - Curated knowledge base on AI memory for LLMs and agents, covering long-term memory, reasoning, retrieval, and system design. by @IAAR-Shanghai (1,208 stars)
  • Awesome-Agent-Memory - Curated systems, benchmarks, and papers on memory for LLMs/MLLMs -- long-term context, retrieval, and reasoning. by @TeleAI-UAGI (629 stars)
  • MemSkill - Learning and evolving memory skills for self-evolving agents. Meta-memory that determines what to extract, remember, and forget. by @ViktorAxelsen (572 stars)
  • TeleMem - High-performance drop-in Mem0 replacement. 19% higher accuracy, 43% fewer tokens, and 2.1x speedup via narrative dynamic extraction. by @TeleAI-UAGI (489 stars)

Agent-to-Agent Protocols

Standards and protocols for inter-agent communication and interoperability.

  • Google A2A - Google's open Agent-to-Agent protocol. Enables agent discovery, secure collaboration, and long-running tasks while preserving agent opacity. by @a2aproject (25,696 stars)
  • mcp-use - The fullstack MCP framework to develop MCP Apps for ChatGPT/Claude and MCP Servers for AI Agents. by @mcp-use (10,595 stars)
  • openagent - Enterprise AI platform with MCP and A2A protocol management, knowledge base, and admin interface. by @the-open-agent (5,608 stars)
  • ViteMCP - A TypeScript framework for building MCP servers. by @punkpeye (3,262 stars)
  • arcade-mcp - MCP server framework and tool-development library for building custom agent capabilities and authenticated tool calls. by @ArcadeAI (1,022 stars)
  • A2A x402 - A2A protocol extension adding x402 on-chain payments, letting agents monetize services over Agent-to-Agent calls. by @google-agentic-commerce (559 stars)
  • GEP MCP Server - MCP Server for Genome Evolution Protocol. Exposes evolution tools to Claude Desktop, Cursor, and any MCP client. by @EvoMap (6 stars)

Agent Development Platforms

Platforms and tools for building, deploying, and managing AI agents.

  • dify - Production-ready platform for building agentic AI workflows with visual orchestration. by @langgenius (155,124 stars)
  • LangChain - Full-stack agent engineering platform with composable chains, tools, and memory integration. by @langchain-ai (145,983 stars)
  • OpenHands - Open platform for AI software developers as generalist agents. Autonomous coding, debugging, and deployment. by @OpenHands (86,970 stars)
  • CowAgent - Super AI assistant based on LLMs with autonomous thinking, task planning, skill creation, and long-term memory. by @zhayujie (46,845 stars)
  • agno - Production-ready agent framework that turns agents into deployable services with multi-framework support. by @agno-agi (42,109 stars)
  • langgraph - Build resilient language agents as stateful graphs with persistence and streaming. by @langchain-ai (41,289 stars)
  • CoPaw - Co Personal Agent Workstation built on AgentScope. Desktop agent platform with multi-agent collaboration and tool integration. by @agentscope-ai (35,081 stars)
  • mastra - TypeScript framework for building AI-powered applications with agent workflows and RAG. by @mastra-ai (27,818 stars)
  • AgenticSeek - Fully local autonomous agent with browsing, coding, and multi-agent capabilities. No API keys required. by @Fosowl (27,188 stars)
  • haystack - Open-source AI orchestration framework for building context-engineered production applications. by @deepset-ai (26,451 stars)
  • Coze Studio - AI agent development platform with visual tools for creating, debugging, and deploying agents. by @coze-dev (21,568 stars)
  • Google ADK - Open-source Python toolkit by Google for building, evaluating, and deploying sophisticated AI agents. by @google (21,458 stars)
  • PydanticAI - Type-safe AI agent framework built on Pydantic with structured outputs and dependency injection. by @pydantic (19,816 stars)
  • Parlant - The conversational control layer for customer-facing AI agents. A context-engineering framework for controlling interactions. by @emcie-co (18,283 stars)
  • OpenFang - Open-source Agent Operating System for deploying and managing AI agents. by @RightNow-AI (18,170 stars)
  • agents - Framework for building realtime voice AI agents with speech-to-speech pipelines. by @livekit (14,080 stars)
  • ten-framework - Open-source framework for building conversational voice AI agents. by @TEN-framework (11,114 stars)
  • Agent-Squad - Flexible framework for managing multiple AI agents and handling complex conversations. by @2FastLabs (7,756 stars)
  • PySpur - Visual playground for agentic workflows with rapid iteration on multi-agent pipelines. by @PySpur-Dev (5,782 stars)
  • MS-Agent - Lightweight framework by ModelScope to empower agentic execution of complex tasks with memory and deep research. by @modelscope (4,379 stars)
  • SandBase Harness - Local-first, self-hosted AI agent runtime with sandboxed sessions, MCP tool governance, persistent memory, credentials, audit and replay, and a local Console. by @sandbaseai (643 stars)

Agent Coding and Software Engineering

AI agents that write, debug, and maintain code autonomously.

  • Claude Code - Terminal-native agentic coding tool from Anthropic. Understands your codebase and executes tasks through natural language. by @anthropics (144,495 stars)
  • Codex - Lightweight coding agent from OpenAI written in Rust. Runs locally as CLI, IDE extension, or desktop app. by @OpenAI (122,629 stars)
  • Pi - Self-extensible coding agent and agent harness. Bundles an interactive coding CLI, an agent runtime with tool calling and state, and a unified multi-provider LLM API. by @earendil-works (103,245 stars)
  • agent-skills - Production-grade engineering skills and best practices for AI coding agents. by @addyosmani (93,070 stars)
  • Taste-Skill - High-Agency frontend skill that helps AI generate less generic, more tasteful outputs. by @Leonxlnx (85,543 stars)
  • Cline - Autonomous coding agent available as an IDE extension, CLI, or SDK. Plans and executes multi-step edits with human-in-the-loop approval. by @cline (67,698 stars)
  • goose - Open-source extensible AI coding agent that goes beyond code suggestions. by @aaif-goose (54,044 stars)
  • Aider - AI pair programming in your terminal. Edit code with LLMs across 100+ languages with deep Git integration. by @Aider-AI (48,837 stars)
  • Qwen Code - Open-source AI coding agent that lives in your terminal, optimized for Qwen-Coder models. by @QwenLM (27,716 stars)
  • context-mode - Context window optimization tool for AI coding agents with sandboxed tool output and 98% token reduction. by @mksglu (21,484 stars)
  • SWE-Agent - Automatically fix GitHub issues and handle cybersecurity challenges. State-of-the-art on SWE-bench. by @SWE-agent (20,287 stars)
  • Devika - The first open-source implementation of an Agentic Software Engineer. An open-source alternative to Devin. by @stitionai (19,564 stars)
  • Plandex - Open-source AI coding agent designed for large projects and complex real-world tasks with persistent context. by @plandex-ai (15,628 stars)
  • Trae Agent - LLM-based agent by ByteDance for general-purpose software engineering tasks. by @bytedance (12,078 stars)
  • Open SWE - Open-source asynchronous coding agent by LangChain for software engineering tasks. by @langchain-ai (10,684 stars)
  • Mini-SWE-Agent - The 100-line AI agent that solves GitHub issues. Radically simple but scores >74% on SWE-bench verified. by @SWE-agent (7,230 stars)
  • Reflexion - Language agents with verbal reinforcement learning. Agents that learn from mistakes through self-reflection. by @noahshinn (3,262 stars)
  • Autohand Code - Self-evolving coding agent that runs in your terminal. by @autohandai (186 stars)

Multi-Agent Frameworks

Multi-agent frameworks and collaborative agent systems that are relevant to agent evolution and infrastructure.

  • MetaGPT - Multi-agent framework that assigns roles to GPTs to form a collaborative software entity. by @FoundationAgents (70,276 stars)
  • autogen - Programming framework for building agentic AI systems with multi-agent conversations. by @microsoft (60,889 stars)
  • crewAI - Framework for orchestrating role-playing autonomous AI agents for collaborative task execution. by @crewAIInc (58,265 stars)
  • Vibe-Trading - Personal trading agent for multi-agent market analysis and backtesting. by @HKUDS (33,076 stars)
  • agentscope - Multi-agent platform for building agents you can see, understand and trust. by @agentscope-ai (31,152 stars)
  • openai-agents-python - Lightweight framework for multi-agent workflows by OpenAI. by @openai (29,285 stars)
  • GenAI Agents - Comprehensive tutorials and implementations covering 50+ generative AI agent techniques from basic to advanced. by @NirDiamant (24,201 stars)
  • swarm - Educational framework exploring ergonomic lightweight multi-agent orchestration by OpenAI. by @openai (21,958 stars)
  • camel - Multi-agent framework for finding the scaling law of agents through role-playing communication. by @camel-ai (17,692 stars)
  • agent-framework - Framework for building, orchestrating and deploying AI agents and multi-agent systems. by @microsoft (13,413 stars)
  • PraisonAI - AI employee team platform that automates and solves complex challenges with multi-agent collaboration. by @MervinPraison (9,032 stars)
  • swarms - Enterprise-grade production-ready multi-agent orchestration framework. by @kyegomez (7,157 stars)
  • MindSearch - LLM-based multi-agent framework for web search engines with deep information seeking. by @InternLM (6,919 stars)
  • open-multi-agent - TypeScript multi-agent framework with one function call from goal to result. by @open-multi-agent (6,894 stars)
  • agency-swarm - Reliable multi-agent orchestration framework for building AI agent swarms. by @VRSEN (4,553 stars)
  • Golutra - Multi-agent AI orchestration platform for automation, workflows, and developer tools. by @golutra (3,830 stars)
  • GoClaw - OpenClaw rebuilt in Go with multi-tenant isolation, 5-layer security, and native concurrency for deploying AI agent teams at scale. by @nextlevelbuilder (3,595 stars)
  • BotSharp - AI multi-agent framework built on .NET for enterprise conversational applications. by @SciSharp (3,100 stars)
  • agentUniverse - LLM multi-agent framework for building, customizing and running collaborative agents. by @agentuniverse-ai (2,351 stars)
  • LLMStack - No-code multi-agent framework for building LLM agent workflows and applications. by @trypromptly (2,310 stars)
  • Sage - Multi-agent system framework for complex tasks. by @ZHangZHengEric (1,217 stars)

Tools and frameworks for automatically optimizing agent prompts, instructions, and behavioral patterns.

  • Promptfoo - Open-source LLM evaluation and red-teaming framework. Test prompts, agents, and RAGs with 90+ model providers and 67+ security plugins. by @promptfoo (24,949 stars)
  • TextGrad - Automatic differentiation via text. Backpropagation through LLM-provided textual gradients, published in Nature. by @zou-group (3,721 stars)

Agent Safety and Guardrails

Projects focused on controlling agent actions, enforcing policies, and preventing harmful behavior.

  • NeMo Guardrails - NVIDIA's toolkit for adding programmable guardrails to LLM conversational systems. Policy-based safety controls. by @NVIDIA-NeMo (7,086 stars)
  • AgentDoG - Diagnostic guardrail framework for AI agent safety and security. Detects and intercepts unsafe agent behavior at runtime. by @AI45Lab (696 stars)

Embodied AI

Projects connecting AI agents to physical devices, robotics, and real-world environments.

  • LeRobot - Open-source robotics framework by Hugging Face. Models, datasets, and tools for real-world robotics in PyTorch. by @huggingface (27,341 stars)
  • Open-AutoGLM - An Open Phone Agent Model and Framework. Unlocking the AI Phone for Everyone. by @zai-org (26,194 stars)
  • Nanobrowser - Chrome extension for AI-powered web automation. Run multi-agent workflows using your own AI keys. by @nanobrowser (13,753 stars)
  • Mobile MCP - Model Context Protocol Server for Mobile Automation and Scraping (iOS, Android, Emulators and Real Devices). by @mobile-next (6,593 stars)
  • XcodeBuildMCP - A MCP server and CLI for agent use when working on iOS and macOS projects. by @getsentry (6,357 stars)
  • agent-device - CLI that lets AI agents drive real iOS and Android devices — taps, text input, screenshots, and app control for mobile automation. by @callstack (4,462 stars)
  • ROS-LLM - Framework for embodied intelligence in ROS. Natural language interactions with LLMs for robot control. by @Auromix (828 stars)
  • RAI - Vendor-agnostic agentic framework for Physical AI and robotics. Connects LLM agents to ROS 2 tools for perception, reasoning, and control. by @RobotecAI (585 stars)

Key Research Papers

Surveys

Self-Evolution and Lifelong Learning

Memory Optimization

Prompt and Behaviour Evolution

Tool and Code Evolution

Reasoning and Planning

Safety, Red-Teaming, and Alignment

Embodied AI and Robotics

Benchmarks and Evaluation

  • SWE-bench (ICLR'24) - Can agents resolve real-world GitHub issues?
  • AgentBench (ICLR'24) - Multi-dimensional evaluation of LLMs as agents.
  • WebArena (ICLR'24) - Realistic web environment for autonomous agents.
  • OSWorld (NeurIPS'24) - Open-ended tasks in real computer environments.
  • GAIA (ICLR'23) - General AI assistant capabilities benchmark.
  • SWE-Milestone (ICML'26) - Evaluating agents on continuous software evolution.
  • LoCoMo (arXiv'25) - Long-context memory benchmark for agent memory systems.
  • ClawBench (arXiv'26) - Live-web evaluation with 281 total tasks (V1: 152; V2: 129), isolated execution, and replayable five-layer traces.
  • ATM-Bench (arXiv'26) - First multimodal, multi-source benchmark for personalized referential memory QA over ~4 years of personal records (emails, images, videos).
  • PerspectiveGap (arXiv'26) - The first benchmark for multi-agent orchestration prompt writing, across 110 scenarios and 10 topologies.

Community and Knowledge

  • Awesome-Self-Evolving-Agents - A comprehensive survey of self-evolving AI agents. Covers single-agent optimization, multi-agent optimization, and domain-specific approaches. by @ANative-Lab (2,474 stars)

Footnotes

Maintained by EvoMap. See contributing guidelines for how to submit a project or paper.

Also check out Awesome Agent Swarm for multi-agent orchestration, swarm intelligence, and collaborative agent systems.

Star History Chart

About

A curated list of AI Agent evolution, memory systems, multi-agent architectures, and self-improvement projects. | evomap.ai

Topics

a2aagent-evolutionagent-frameworkagent-protocolagentic-aiai-agentautonomous-agentawesomeawesome-listevomapllmllm-agentmcpmemory-systemmulti-agentprompt-engineeringself-evolvingself-improvementskill-libraryskills

Resources

Readme
License

Contributing

Contributing
Activity
Custom properties

Stars

222 stars

Watchers

1 watching

Forks

39 forks
Report repository

Releases

Packages

Contributors

Languages