A curated collection of projects, papers, and resources advancing AI Agent self-evolution -- from memory and reasoning optimization to multi-agent collaboration and autonomous self-improvement.
Maintained by EvoMap -- the open platform for Agent evolution and collaboration.
graph LR
Root["Agent Evolution"] --> SingleAgent["Single-Agent<br/>Optimization"]
Root --> MultiAgent["Multi-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"]
MultiAgent --> Frameworks["Multi-Agent<br/>Frameworks"]
MultiAgent --> WorkflowOpt["Workflow<br/>Optimization"]
Infrastructure --> Protocols["A2A & MCP<br/>Protocols"]
Infrastructure --> Platforms["Agent Development<br/>Platforms"]
Infrastructure --> Embodied["Embodied AI<br/>& Device Control"]
Open-Source Projects
- 1. Agent Evolution and Self-Improvement
- 2. Memory Systems
- 3. Multi-Agent Frameworks
- 4. Agent-to-Agent Protocols
- 5. Agent Development Platforms
- 6. Embodied AI
Research
Community
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 (17711 stars)
autonomousagent-osevolution - SuperAGI -- A dev-first open source autonomous AI agent framework. Enabling developers to build, manage and run useful autonomous agents. by @TransformerOptimus (17238 stars)
autonomousself-improvementdev-tools - Agent Zero -- Agent Zero AI framework -- a general-purpose AI agent that learns and evolves through interaction. by @agent0ai (15840 stars)
autonomousgeneral-purposelearning - Agents (aiwaves) -- An open-source framework for data-centric, self-evolving autonomous language agents. by @aiwaves-cn (5878 stars)
self-evolvingautonomousdata-centric - evolver -- The GEP-powered self-evolution engine for AI agents. Genome Evolution Protocol enables agents to evolve autonomously. by @autogame-17 (1316 stars)
gepself-evolutiongenome - EvoAgentX -- An automated framework for evolving agentic workflows. Optimizes agent prompts, tools, and pipelines via evolutionary algorithms. by @EvoAgentX (1200 stars)
workflow-evolutionprompt-optimizationautomated - Darwin Godel Machine -- Open-ended evolution of self-improving agents. Agents that rewrite their own code to improve performance. by @jennyzzt (800 stars)
self-improvingcode-rewritingopen-ended - OpenEvolve -- An open-source evolutionary coding agent inspired by AlphaEvolve. Evolves code solutions through LLM-driven mutation and selection. by @codelion (600 stars)
evolutionary-codingalphaevolveopen-source - Alita -- Generalist Agent enabling scalable agentic reasoning with minimal predefinition and maximal self-evolution. by @alita-ai (400 stars)
generalistself-evolutionscalable - SEAgent -- Self-Evolving Computer Use Agent with Autonomous Learning from Experience. by @SunzeY (231 stars)
self-evolvingcomputer-useautonomous-learning
Vector, graph, episodic, and hybrid memory architectures for persistent agent cognition.
- Mem0 -- Production-ready AI agent memory with scalable long-term memory. The memory layer for personalized AI. by @mem0ai (25000 stars)
production-readyscalablepersonalized - Memvid -- Memory layer for AI Agents. Replace complex RAG pipelines with a serverless, single-file memory layer. by @memvid (13278 stars)
memory-layerserverlessretrieval - memU -- Memory system for 24/7 proactive agents. Persistent memory across sessions and platforms. by @NevaMind-AI (12654 stars)
proactive-agentspersistentcross-platform - EverMemOS -- Long-term memory for 24/7 AI agents across LLMs and platforms. by @EverMind-AI (2403 stars)
long-term-memorycross-platformpersistent - A-MEM -- Agentic Memory for LLM Agents. Self-organizing memory that autonomously manages what to remember and forget. by @agentic-memory (1500 stars)
agenticself-organizingautonomous - mcp-memory-service -- Open-source persistent memory for AI agent pipelines. REST API + knowledge graph + autonomous consolidation. by @doobidoo (1463 stars)
mcpknowledge-graphpersistent-memory - Awesome-AI-Memory -- A curated knowledge base on AI memory for LLMs and agents, covering long-term memory, reasoning, retrieval, and system design. by @IAAR-Shanghai (454 stars)
surveyknowledge-baseresearch - TeleMem -- High-performance drop-in replacement for Mem0, featuring semantic deduplication, long-term dialogue memory, and multimodal video reasoning. by @TeleAI-UAGI (440 stars)
semantic-dedupmultimodalhigh-performance - Aetherius -- A private, locally-operated AI Assistant with realistic Long Term Memory and thought formation using Open Source LLMs. by @libraryofcelsus (313 stars)
locallong-term-memoryopen-source - nocturne_memory -- A lightweight, rollbackable, and visual Long-Term Memory Server for MCP Agents with graph-like structured memory. by @Dataojitori (294 stars)
graph-memoryrollbackablemcp - Awesome-Agent-Memory -- Curated systems, benchmarks, and papers on memory for LLMs/MLLMs -- long-term context, retrieval, and reasoning. by @TeleAI-UAGI (262 stars)
surveybenchmarksresearch
Frameworks for orchestrating multiple AI agents to collaborate on complex tasks.
- MetaGPT -- The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming. by @FoundationAgents (64830 stars)
software-companyrole-playingnlp - AutoGen -- A programming framework for agentic AI. Build multi-agent applications with conversational patterns. by @microsoft (55254 stars)
conversationalmicrosoftagentic - CrewAI -- Framework for orchestrating role-playing, autonomous AI agents. Collaborative intelligence for complex tasks. by @crewAIInc (45322 stars)
role-playingcollaborativeorchestration - DSPy -- Compiling declarative language model calls into state-of-the-art pipelines. Programming -- not prompting -- foundation models. by @stanfordnlp (25000 stars)
compilerpipelinestanford - Haystack -- Open-source AI orchestration framework for building production-ready LLM applications with modular pipelines and agent workflows. by @deepset-ai (24418 stars)
pipelinesragproduction-ready - Mastra -- From the team behind Gatsby. A framework for building AI-powered applications and agents with modern TypeScript. by @mastra-ai (21767 stars)
typescriptmoderngatsby-team - Swarm (OpenAI) -- Educational framework exploring ergonomic, lightweight multi-agent orchestration. by @openai (21090 stars)
lightweighteducationalopenai - OpenAI Agents Python -- A lightweight, powerful framework for multi-agent workflows from OpenAI. by @openai (19388 stars)
pythonlightweightopenai - DB-GPT -- AI Native Data App Development framework with AWEL (Agentic Workflow Expression Language) and Agents. by @eosphoros-ai (18213 stars)
data-appworkflowawel - CAMEL -- The first and the best multi-agent framework. Finding the Scaling Law of Agents. by @camel-ai (16203 stars)
scaling-lawresearchmulti-agent - Microsoft Agent Framework -- A framework for building, orchestrating and deploying AI agents with support for Python and .NET. by @microsoft (7709 stars)
microsoftpythondotnet - Agent Squad -- Flexible and powerful framework for managing multiple AI agents and handling complex conversations. by @awslabs (7491 stars)
awsflexibleconversations - MindSearch -- An LLM-based Multi-agent Framework of Web Search Engine (like Perplexity.ai Pro and SearchGPT). by @InternLM (6788 stars)
web-searchperplexity-likeresearch - Swarms -- The Enterprise-Grade Production-Ready Multi-Agent Orchestration Framework. by @kyegomez (5845 stars)
enterpriseorchestrationproduction-ready - PraisonAI -- Production-ready Multi AI Agents framework. Low-code solution for multi-agent LLM systems. by @MervinPraison (5635 stars)
low-codeproduction-readycustomizable - Agency Swarm -- Reliable Multi-Agent Orchestration Framework for building agent teams. by @VRSEN (4026 stars)
orchestrationreliableagent-teams - AFlow -- Automating agentic workflow generation. Automatically designs optimal multi-agent workflows for given tasks. by @geekan (1500 stars)
workflow-automationauto-designoptimization
Standards and protocols for inter-agent communication and interoperability.
- mcp-use -- The fullstack MCP framework to develop MCP Apps for ChatGPT/Claude and MCP Servers for AI Agents. by @mcp-use (9383 stars)
mcpfullstackchatgpt - Casibase -- AI Cloud OS: Enterprise-level AI knowledge base and MCP/A2A management platform with admin UI. by @casibase (4456 stars)
a2amcpenterprise - Python A2A -- A powerful library for implementing Google's Agent-to-Agent (A2A) protocol for seamless inter-agent communication. by @themanojdesai (981 stars)
a2agooglepython - ATLAS MCP Server -- Neo4j-powered task management system for LLM Agents with three-tier architecture (Projects, Tasks, Knowledge). by @cyanheads (467 stars)
mcpneo4jtask-management - A2A x402 -- The A2A x402 Extension brings cryptocurrency payments to the Agent-to-Agent protocol, enabling agents to monetize their services. by @google-agentic-commerce (462 stars)
a2apaymentscrypto - Coral Anemoi -- A Semi-Centralized Multi-agent System based on Agent-to-Agent Communication MCP server. by @Coral-Protocol (373 stars)
a2amcpcoral - A2A Go -- Golang SDK for the A2A Protocol. by @a2aproject (270 stars)
a2agolangsdk - Mangaba AI -- Minimal framework for creating AI agents with A2A and MCP protocols. by @Mangaba-ai (184 stars)
a2amcpminimal - A2A MCP Server -- Bridges Model Context Protocol with Agent-to-Agent protocol, enabling MCP-compatible assistants to interact with A2A agents. by @GongRzhe (143 stars)
a2amcpbridge
Platforms and tools for building, deploying, and managing AI agents.
- OpenHands -- An open platform for AI software developers as generalist agents. Autonomous coding, debugging, and deployment. by @All-Hands-AI (50000 stars)
coding-agentautonomousgeneralist - Parlant -- The conversational control layer for customer-facing AI agents. A context-engineering framework for controlling interactions. by @emcie-co (17807 stars)
conversationalcustomer-facingcontrol-layer - OpenFang -- Open-source Agent Operating System for deploying and managing AI agents. by @RightNow-AI (11641 stars)
agent-osopen-sourcedeployment - TEN Framework -- Open-source framework for conversational voice AI agents. by @TEN-framework (10217 stars)
voiceconversationalreal-time - LiveKit Agents -- A framework for building realtime voice AI agents. by @livekit (9604 stars)
voicerealtimelivekit - BotSharp -- AI Multi-Agent Framework in .NET for building intelligent agents. by @SciSharp (3022 stars)
dotnetcsharpmulti-agent - LLMStack -- No-code multi-agent framework to build LLM Agents, workflows and applications with your data. by @trypromptly (2294 stars)
no-codeworkflowsdata-integration - agentUniverse -- A LLM multi-agent framework that allows developers to easily build multi-agent applications. by @agentuniverse-ai (2129 stars)
developer-friendlymulti-agentapplications
Projects connecting AI agents to physical devices, robotics, and real-world environments.
- Open-AutoGLM -- An Open Phone Agent Model and Framework. Unlocking the AI Phone for Everyone. by @zai-org (24074 stars)
phonemobileautomate - XcodeBuildMCP -- A MCP server and CLI for agent use when working on iOS and macOS projects. by @getsentry (4585 stars)
xcodeiosmacos - Mobile MCP -- Model Context Protocol Server for Mobile Automation and Scraping (iOS, Android, Emulators and Real Devices). by @mobile-next (3727 stars)
mobileautomationmcp - ROS-LLM -- Framework for embodied intelligence in ROS. Natural language interactions with LLMs for robot control. by @Auromix (743 stars)
rosroboticsnatural-language
Selected papers that shaped the field of agent evolution and self-improvement.
| Paper | Venue | Key Contribution |
|---|---|---|
| A Comprehensive Survey of Self-Evolving AI Agents | arXiv'25 | Taxonomy of single-agent, multi-agent, and domain-specific evolution |
| Symbolic Learning Enables Self-Evolving Agents | arXiv'24 | Agents that evolve through symbolic representation learning |
| Building Self-Evolving Agents via Experience-Driven Lifelong Learning | arXiv'25 | Framework and benchmark for lifelong agent learning |
| Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents | arXiv'25 | Agents that rewrite their own code through evolutionary pressure |
| EvoAgent: Self-evolving Agent with Continual World Model | arXiv'25 | Continual world model for long-horizon task evolution |
| Absolute Zero: Reinforced Self-play Reasoning with Zero Data | arXiv'25 | Self-play reasoning without any training data |
| Paper | Venue | Key Contribution |
|---|---|---|
| Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory | arXiv'25 | Production architecture for scalable agent memory |
| A-MEM: Agentic Memory for LLM Agents | arXiv'25 | Self-organizing memory with autonomous management |
| Agent Workflow Memory | ICML'24 | Memory tied to agent workflow patterns |
| MemoryBank: Enhancing Large Language Models with Long-Term Memory | AAAI'24 | Structured long-term memory for LLMs |
| Compress to Impress: Compressive Memory in Real-World Long-Term Conversations | ICLR'25 | Compression-based memory for extended dialogues |
| Paper | Venue | Key Contribution |
|---|---|---|
| EvoPrompt: Connecting LLMs with Evolutionary Algorithms | ICLR'24 | Evolutionary algorithms for prompt optimization |
| Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution | ICML'24 | Prompts that evolve themselves recursively |
| Large Language Models as Optimizers (OPRO) | ICLR'24 | Using LLMs to optimize their own prompts |
| TextGrad: Automatic Differentiation via Text | arXiv'24 | Gradient-like optimization through text feedback |
| Paper | Venue | Key Contribution |
|---|---|---|
| Self-Evolving Multi-Agent Collaboration Networks for Software Development | ICLR'25 | Multi-agent systems that evolve their collaboration patterns |
| AFlow: Automating Agentic Workflow Generation | ICLR'25 | Automated design of multi-agent workflows |
| Automated Design of Agentic Systems (ADAS) | ICLR'25 | Meta-learning for automatic agent system design |
| GPTSwarm: Language Agents as Optimizable Graphs | ICML'24 | Graph-based optimization of agent collaboration |
| AgentVerse: Facilitating Multi-Agent Collaboration | ICLR'24 | Emergent behaviors in multi-agent environments |
| Paper | Venue | Key Contribution |
|---|---|---|
| AlphaEvolve: A coding agent for scientific and algorithmic discovery | Google'25 | LLM-driven evolutionary code improvement |
| Learning Evolving Tools for Large Language Models | ICLR'25 | Tools that co-evolve with agent capabilities |
| CREATOR: Tool Creation for Disentangling Abstract and Concrete Reasoning | EMNLP'23 | Agents that create their own tools |
| ToolRL: Reward is All Tool Learning Needs | arXiv'25 | Reinforcement learning for tool use optimization |
| Resource | Description |
|---|---|
| Awesome-Self-Evolving-Agents by @EvoAgentX | Comprehensive survey covering single-agent, multi-agent, and domain-specific evolution (1.9k stars) |
| Awesome-AI-Memory by @IAAR-Shanghai | Curated knowledge base on AI memory for LLMs and agents (454 stars) |
| Awesome-Agent-Memory by @TeleAI-UAGI | Systems, benchmarks, and papers on memory for LLMs/MLLMs (262 stars) |
| Benchmark | Paper | What It Measures |
|---|---|---|
| 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 |
- Learn Agentic AI -- Learn Agentic AI using DACA Design Pattern: OpenAI Agents SDK, Memory, MCP, A2A, Knowledge Graphs, Dapr, and Kubernetes. by @panaversity (3945 stars)
educationa2amcp - Awesome-Self-Evolving-Agents -- A comprehensive survey of self-evolving AI agents. Covers single-agent optimization, multi-agent optimization, and domain-specific approaches. by @EvoAgentX (1900 stars)
surveyresearchcomprehensive - OpenClaw QA -- OpenClaw developer community -- Q&A, real-world practices, AI Agent memory systems, multi-agent architectures, and evolution. by @YuLin807 (58 stars)
openclawcommunityqa
See CONTRIBUTING.md for guidelines on submitting a project or paper.
You can also open an issue to suggest a project for inclusion.
This list is maintained by EvoMap. Our flagship project evolver implements the Genome Evolution Protocol (GEP) -- a self-evolution engine for AI agents that treats agent behavior as an evolving genome.
| Resource | Link |
|---|---|
| evolver | github.com/EvoMap/evolver |
| evomap.ai | evomap.ai |
| Discussions | evolver Discussions |
CC BY-SA 4.0