{"id":133401,"url":"https://github.com/EvoMap/awesome-agent-evolution","name":"awesome-agent-evolution","description":"A curated list of AI Agent evolution, memory systems, multi-agent architectures, and self-improvement projects. | evomap.ai","projects_count":144,"last_synced_at":"2026-06-24T00:00:46.273Z","repository":{"id":342739068,"uuid":"1174978748","full_name":"EvoMap/awesome-agent-evolution","owner":"EvoMap","description":"A curated list of AI Agent evolution, memory systems, multi-agent architectures, and self-improvement projects. | evomap.ai","archived":false,"fork":false,"pushed_at":"2026-06-11T04:08:27.000Z","size":329,"stargazers_count":136,"open_issues_count":7,"forks_count":20,"subscribers_count":2,"default_branch":"main","last_synced_at":"2026-06-11T06:34:23.126Z","etag":null,"topics":["a2a","agent-evolution","agent-framework","agent-protocol","agentic-ai","ai-agent","autonomous-agent","awesome","awesome-list","evomap","llm","llm-agent","mcp","memory-system","multi-agent","prompt-engineering","self-evolving","self-improvement","skill-library","skills"],"latest_commit_sha":null,"homepage":null,"language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/EvoMap.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-03-07T04:11:34.000Z","updated_at":"2026-06-10T09:37:22.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/EvoMap/awesome-agent-evolution","commit_stats":null,"previous_names":["evomap/awesome-agent-evolution"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/EvoMap/awesome-agent-evolution","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EvoMap%2Fawesome-agent-evolution","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EvoMap%2Fawesome-agent-evolution/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EvoMap%2Fawesome-agent-evolution/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EvoMap%2Fawesome-agent-evolution/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/EvoMap","download_url":"https://codeload.github.com/EvoMap/awesome-agent-evolution/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EvoMap%2Fawesome-agent-evolution/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34711203,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-23T02:00:07.161Z","response_time":65,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"created_at":"2026-06-07T04:30:12.463Z","updated_at":"2026-06-24T00:00:46.273Z","primary_language":null,"list_of_lists":false,"displayable":true,"categories":["Key Research Papers","Agent Coding and Software Engineering","Agent Development Platforms","Prompt and Behaviour Optimization","Agent Evolution and Self-Improvement","Memory Systems","Benchmarks and Evaluation","Embodied AI","Agent-to-Agent Protocols","Agent Safety and Guardrails","Community and Knowledge","Footnotes"],"sub_categories":["Reasoning and Planning","Tool and Code Evolution","Prompt and Behaviour Evolution","Memory Optimization","Embodied AI and Robotics","Self-Evolution and Lifelong Learning","Surveys","Safety, Red-Teaming, and Alignment"],"readme":"# Awesome Agent Evolution [![Awesome](https://awesome.re/badge.svg)](https://awesome.re)\n\n\u003e AI Agent self-evolution, memory systems, autonomous self-improvement, and the infrastructure that powers them.\n\n## Contents\n\n- [Taxonomy](#taxonomy)\n- [Agent Evolution and Self-Improvement](#agent-evolution-and-self-improvement)\n- [Memory Systems](#memory-systems)\n- [Agent-to-Agent Protocols](#agent-to-agent-protocols)\n- [Agent Development Platforms](#agent-development-platforms)\n- [Agent Coding and Software Engineering](#agent-coding-and-software-engineering)\n- [Prompt and Behaviour Optimization](#prompt-and-behaviour-optimization)\n- [Agent Safety and Guardrails](#agent-safety-and-guardrails)\n- [Embodied AI](#embodied-ai)\n- [Key Research Papers](#key-research-papers)\n- [Benchmarks and Evaluation](#benchmarks-and-evaluation)\n- [Community and Knowledge](#community-and-knowledge)\n\n## Taxonomy\n\n```mermaid\ngraph LR\n    Root[\"Agent Evolution\"] --\u003e SingleAgent[\"Single-Agent\u003cbr/\u003eOptimization\"]\n    Root --\u003e Infrastructure[\"Infrastructure\u003cbr/\u003e\u0026 Protocols\"]\n\n    SingleAgent --\u003e Evolution[\"Self-Evolution\u003cbr/\u003e\u0026 Self-Improvement\"]\n    SingleAgent --\u003e Memory[\"Memory\u003cbr/\u003eSystems\"]\n    SingleAgent --\u003e PromptOpt[\"Prompt \u0026 Behaviour\u003cbr/\u003eOptimization\"]\n\n    Infrastructure --\u003e Protocols[\"A2A \u0026 MCP\u003cbr/\u003eProtocols\"]\n    Infrastructure --\u003e Platforms[\"Agent Development\u003cbr/\u003ePlatforms\"]\n    Infrastructure --\u003e Coding[\"Agent Coding \u0026\u003cbr/\u003eSoftware Engineering\"]\n    Infrastructure --\u003e Safety[\"Agent Safety\u003cbr/\u003e\u0026 Guardrails\"]\n    Infrastructure --\u003e Embodied[\"Embodied AI\u003cbr/\u003e\u0026 Device Control\"]\n```\n\n## Agent Evolution and Self-Improvement\n\nProjects focused on enabling AI agents to evolve, learn, and improve autonomously.\n\n\u003c!-- AUTOGEN:evolution --\u003e\n- [**Eliza**](https://github.com/elizaOS/eliza) - Autonomous agents for everyone. A framework for creating and deploying AI agents that evolve over time. by [@elizaOS](https://github.com/elizaOS) (18,598 stars)\n- [**Agent Zero**](https://github.com/agent0ai/agent-zero) - General-purpose AI agent framework that learns and evolves through interaction. by [@agent0ai](https://github.com/agent0ai) (18,108 stars)\n- [**SuperAGI**](https://github.com/TransformerOptimus/SuperAGI) - A dev-first open source autonomous AI agent framework. Build, manage and run self-improving autonomous agents. by [@TransformerOptimus](https://github.com/TransformerOptimus) (17,573 stars)\n- [**evolver**](https://github.com/EvoMap/evolver) - The GEP-powered self-evolution engine for AI agents. Genome Evolution Protocol enables agents to evolve autonomously via mutation and selection. by [@EvoMap](https://github.com/EvoMap) (8,756 stars)\n- [**OpenEvolve**](https://github.com/algorithmicsuperintelligence/openevolve) - Open-source evolutionary coding agent inspired by AlphaEvolve. Evolves code solutions through LLM-driven mutation and selection. by [@algorithmicsuperintelligence](https://github.com/algorithmicsuperintelligence) (6,564 stars)\n- [**Agents (aiwaves)**](https://github.com/aiwaves-cn/agents) - An open-source framework for data-centric, self-evolving autonomous language agents. by [@aiwaves-cn](https://github.com/aiwaves-cn) (5,932 stars)\n- [**EvoAgentX**](https://github.com/EvoAgentX/EvoAgentX) - Automated framework for evolving agentic workflows. Optimizes agent prompts, tools, and pipelines via evolutionary algorithms. by [@EvoAgentX](https://github.com/EvoAgentX) (3,077 stars)\n- [**HyperAgents**](https://github.com/facebookresearch/HyperAgents) - Self-referential self-improving agents by Meta. DGM-Hyperagents add an optimization layer so agents edit their own improvement process. by [@facebookresearch](https://github.com/facebookresearch) (2,583 stars)\n- [**SIA**](https://github.com/hexo-ai/sia) - Self-improving AI framework that autonomously optimizes the performance of any AI system through iterative evaluation and refinement. by [@hexo-ai](https://github.com/hexo-ai) (1,753 stars)\n- [**Agent0**](https://github.com/aiming-lab/Agent0) - Self-evolving agent framework from UNC/Salesforce/Stanford. Improves without human-curated datasets via curriculum and executor agent competition. by [@aiming-lab](https://github.com/aiming-lab) (1,220 stars)\n- [**Ouroboros**](https://github.com/razzant/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](https://github.com/razzant) (642 stars)\n- [**A-Evolve**](https://github.com/A-EVO-Lab/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](https://github.com/A-EVO-Lab) (611 stars)\n- [**SEAgent**](https://github.com/SunzeY/SEAgent) - Self-Evolving Computer Use Agent with Autonomous Learning from Experience. by [@SunzeY](https://github.com/SunzeY) (251 stars)\n\u003c!-- /AUTOGEN:evolution --\u003e\n\n## Memory Systems\n\nVector, graph, episodic, and hybrid memory architectures for persistent agent cognition.\n\n\u003c!-- AUTOGEN:memory --\u003e\n- [**Mem0**](https://github.com/mem0ai/mem0) - Production-ready AI agent memory with scalable long-term memory. 26% improvement over baseline on LOCOMO benchmark with 91% latency reduction. by [@mem0ai](https://github.com/mem0ai) (58,775 stars)\n- [**Letta**](https://github.com/letta-ai/letta) - Platform for building stateful agents with advanced self-editing memory. Formerly MemGPT. by [@letta-ai](https://github.com/letta-ai) (23,376 stars)\n- [**agentmemory**](https://github.com/rohitg00/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](https://github.com/rohitg00) (23,222 stars)\n- [**Cognee**](https://github.com/topoteretes/cognee) - Knowledge engine for AI agent memory. Build and query knowledge graphs from unstructured data in 6 lines of code. by [@topoteretes](https://github.com/topoteretes) (17,870 stars)\n- [**Memvid**](https://github.com/memvid/memvid) - Single-file memory layer for AI Agents in Rust. +35% SOTA on LoCoMo with ultra-low latency (0.025ms P50). by [@memvid](https://github.com/memvid) (15,662 stars)\n- [**memU**](https://github.com/NevaMind-AI/memU) - Memory system for 24/7 proactive agents. Persistent memory across sessions and platforms. by [@NevaMind-AI](https://github.com/NevaMind-AI) (13,882 stars)\n- [**EverMemOS**](https://github.com/EverMind-AI/EverOS) - Long-term memory for 24/7 AI agents across LLMs and platforms. by [@EverMind-AI](https://github.com/EverMind-AI) (7,629 stars)\n- [**ChatLab**](https://github.com/ChatLab/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](https://github.com/ChatLab) (6,716 stars)\n- [**TencentDB Agent Memory**](https://github.com/TencentCloud/TencentDB-Agent-Memory) - Fully local long-term memory for AI agents via a four-tier progressive storage architecture, from Tencent Cloud. by [@TencentCloud](https://github.com/TencentCloud) (5,859 stars)\n- [**holaOS**](https://github.com/holaboss-ai/holaOS) - Agent environment for long-horizon work, continuity, and self-evolution. by [@holaboss-ai](https://github.com/holaboss-ai) (5,513 stars)\n- [**honcho**](https://github.com/plastic-labs/honcho) - Memory library for building stateful agents with user context management. by [@plastic-labs](https://github.com/plastic-labs) (5,231 stars)\n- [**memgraph**](https://github.com/memgraph/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](https://github.com/memgraph) (4,164 stars)\n- [**Acontext**](https://github.com/memodb-io/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](https://github.com/memodb-io) (3,538 stars)\n- [**MemMachine**](https://github.com/MemMachine/MemMachine) - Universal memory layer for AI agents. Episodic (graph-based), profile (SQL), and working memory with scalable storage and retrieval. by [@MemMachine](https://github.com/MemMachine) (3,119 stars)\n- [**ReMe**](https://github.com/agentscope-ai/ReMe) - Memory management kit for agents. File-based and vector-based memory systems. SOTA on LoCoMo and HaluMem benchmarks. by [@agentscope-ai](https://github.com/agentscope-ai) (3,094 stars)\n- [**datachain**](https://github.com/datachain-ai/datachain) - Operational data context layer for AI agents providing typed and versioned datasets over multimodal content. by [@datachain-ai](https://github.com/datachain-ai) (2,784 stars)\n- [**nocturne_memory**](https://github.com/Dataojitori/nocturne_memory) - Lightweight, rollbackable Long-Term Memory Server for MCP Agents with graph-like structured memory. by [@Dataojitori](https://github.com/Dataojitori) (1,210 stars)\n- [**Mem9**](https://github.com/mem9-ai/mem9) - Unlimited persistent memory layer for AI agents. Cloud-synced memory across sessions and tools. by [@mem9-ai](https://github.com/mem9-ai) (1,143 stars)\n- [**Awesome-AI-Memory**](https://github.com/IAAR-Shanghai/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](https://github.com/IAAR-Shanghai) (994 stars)\n- [**MemSkill**](https://github.com/ViktorAxelsen/MemSkill) - Learning and evolving memory skills for self-evolving agents. Meta-memory that determines what to extract, remember, and forget. by [@ViktorAxelsen](https://github.com/ViktorAxelsen) (522 stars)\n- [**Awesome-Agent-Memory**](https://github.com/TeleAI-UAGI/Awesome-Agent-Memory) - Curated systems, benchmarks, and papers on memory for LLMs/MLLMs -- long-term context, retrieval, and reasoning. by [@TeleAI-UAGI](https://github.com/TeleAI-UAGI) (476 stars)\n- [**TeleMem**](https://github.com/TeleAI-UAGI/telemem) - High-performance drop-in Mem0 replacement. 19% higher accuracy, 43% fewer tokens, and 2.1x speedup via narrative dynamic extraction. by [@TeleAI-UAGI](https://github.com/TeleAI-UAGI) (466 stars)\n\u003c!-- /AUTOGEN:memory --\u003e\n\n## Agent-to-Agent Protocols\n\nStandards and protocols for inter-agent communication and interoperability.\n\n\u003c!-- AUTOGEN:protocols --\u003e\n- [**Google A2A**](https://github.com/a2aproject/A2A) - Google's open Agent-to-Agent protocol. Enables agent discovery, secure collaboration, and long-running tasks while preserving agent opacity. by [@a2aproject](https://github.com/a2aproject) (24,321 stars)\n- [**mcp-use**](https://github.com/mcp-use/mcp-use) - The fullstack MCP framework to develop MCP Apps for ChatGPT/Claude and MCP Servers for AI Agents. by [@mcp-use](https://github.com/mcp-use) (10,114 stars)\n- [**openagent**](https://github.com/the-open-agent/openagent) - Enterprise AI platform with MCP and A2A protocol management, knowledge base, and admin interface. by [@the-open-agent](https://github.com/the-open-agent) (5,266 stars)\n- [**ViteMCP**](https://github.com/punkpeye/fastmcp) - A TypeScript framework for building MCP servers. by [@punkpeye](https://github.com/punkpeye) (3,198 stars)\n- [**arcade-mcp**](https://github.com/ArcadeAI/arcade-mcp) - MCP server framework and tool-development library for building custom agent capabilities and authenticated tool calls. by [@ArcadeAI](https://github.com/ArcadeAI) (925 stars)\n- [**A2A x402**](https://github.com/google-agentic-commerce/a2a-x402) - A2A protocol extension adding x402 on-chain payments, letting agents monetize services over Agent-to-Agent calls. by [@google-agentic-commerce](https://github.com/google-agentic-commerce) (526 stars)\n- [**GEP MCP Server**](https://github.com/EvoMap/gep-mcp-server) - MCP Server for Genome Evolution Protocol. Exposes evolution tools to Claude Desktop, Cursor, and any MCP client. by [@EvoMap](https://github.com/EvoMap) (4 stars)\n\u003c!-- /AUTOGEN:protocols --\u003e\n\n## Agent Development Platforms\n\nPlatforms and tools for building, deploying, and managing AI agents.\n\n\u003c!-- AUTOGEN:platforms --\u003e\n- [**dify**](https://github.com/langgenius/dify) - Production-ready platform for building agentic AI workflows with visual orchestration. by [@langgenius](https://github.com/langgenius) (145,594 stars)\n- [**LangChain**](https://github.com/langchain-ai/langchain) - Full-stack agent engineering platform with composable chains, tools, and memory integration. by [@langchain-ai](https://github.com/langchain-ai) (139,540 stars)\n- [**OpenHands**](https://github.com/OpenHands/OpenHands) - Open platform for AI software developers as generalist agents. Autonomous coding, debugging, and deployment. by [@OpenHands](https://github.com/OpenHands) (77,481 stars)\n- [**CowAgent**](https://github.com/zhayujie/CowAgent) - Super AI assistant based on LLMs with autonomous thinking, task planning, skill creation, and long-term memory. by [@zhayujie](https://github.com/zhayujie) (45,371 stars)\n- [**agno**](https://github.com/agno-agi/agno) - Production-ready agent framework that turns agents into deployable services with multi-framework support. by [@agno-agi](https://github.com/agno-agi) (40,741 stars)\n- [**langgraph**](https://github.com/langchain-ai/langgraph) - Build resilient language agents as stateful graphs with persistence and streaming. by [@langchain-ai](https://github.com/langchain-ai) (35,007 stars)\n- [**AgenticSeek**](https://github.com/Fosowl/agenticSeek) - Fully local autonomous agent with browsing, coding, and multi-agent capabilities. No API keys required. by [@Fosowl](https://github.com/Fosowl) (26,532 stars)\n- [**haystack**](https://github.com/deepset-ai/haystack) - Open-source AI orchestration framework for building context-engineered production applications. by [@deepset-ai](https://github.com/deepset-ai) (25,589 stars)\n- [**mastra**](https://github.com/mastra-ai/mastra) - TypeScript framework for building AI-powered applications with agent workflows and RAG. by [@mastra-ai](https://github.com/mastra-ai) (25,162 stars)\n- [**Coze Studio**](https://github.com/coze-dev/coze-studio) - AI agent development platform with visual tools for creating, debugging, and deploying agents. by [@coze-dev](https://github.com/coze-dev) (21,000 stars)\n- [**Google ADK**](https://github.com/google/adk-python) - Open-source Python toolkit by Google for building, evaluating, and deploying sophisticated AI agents. by [@google](https://github.com/google) (20,149 stars)\n- [**CoPaw**](https://github.com/agentscope-ai/QwenPaw) - Co Personal Agent Workstation built on AgentScope. Desktop agent platform with multi-agent collaboration and tool integration. by [@agentscope-ai](https://github.com/agentscope-ai) (18,676 stars)\n- [**Parlant**](https://github.com/emcie-co/parlant) - The conversational control layer for customer-facing AI agents. A context-engineering framework for controlling interactions. by [@emcie-co](https://github.com/emcie-co) (18,118 stars)\n- [**OpenFang**](https://github.com/RightNow-AI/openfang) - Open-source Agent Operating System for deploying and managing AI agents. by [@RightNow-AI](https://github.com/RightNow-AI) (17,849 stars)\n- [**PydanticAI**](https://github.com/pydantic/pydantic-ai) - Type-safe AI agent framework built on Pydantic with structured outputs and dependency injection. by [@pydantic](https://github.com/pydantic) (17,805 stars)\n- [**agents**](https://github.com/livekit/agents) - Framework for building realtime voice AI agents with speech-to-speech pipelines. by [@livekit](https://github.com/livekit) (11,012 stars)\n- [**ten-framework**](https://github.com/TEN-framework/ten-framework) - Open-source framework for building conversational voice AI agents. by [@TEN-framework](https://github.com/TEN-framework) (10,680 stars)\n- [**Agent-Squad**](https://github.com/2FastLabs/agent-squad) - Flexible framework for managing multiple AI agents and handling complex conversations. by [@2FastLabs](https://github.com/2FastLabs) (7,661 stars)\n- [**PySpur**](https://github.com/PySpur-Dev/pyspur) - Visual playground for agentic workflows with rapid iteration on multi-agent pipelines. by [@PySpur-Dev](https://github.com/PySpur-Dev) (5,737 stars)\n- [**MS-Agent**](https://github.com/modelscope/ms-agent) - Lightweight framework by ModelScope to empower agentic execution of complex tasks with memory and deep research. by [@modelscope](https://github.com/modelscope) (4,307 stars)\n\u003c!-- /AUTOGEN:platforms --\u003e\n\n## Agent Coding and Software Engineering\n\nAI agents that write, debug, and maintain code autonomously.\n\n\u003c!-- AUTOGEN:coding --\u003e\n- [**Claude Code**](https://github.com/anthropics/claude-code) - Terminal-native agentic coding tool from Anthropic. Understands your codebase and executes tasks through natural language. by [@anthropics](https://github.com/anthropics) (132,937 stars)\n- [**Codex**](https://github.com/openai/codex) - Lightweight coding agent from OpenAI written in Rust. Runs locally as CLI, IDE extension, or desktop app. by [@OpenAI](https://github.com/OpenAI) (91,654 stars)\n- [**Pi**](https://github.com/earendil-works/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](https://github.com/earendil-works) (63,438 stars)\n- [**Cline**](https://github.com/cline/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](https://github.com/cline) (63,419 stars)\n- [**agent-skills**](https://github.com/addyosmani/agent-skills) - Production-grade engineering skills and best practices for AI coding agents. by [@addyosmani](https://github.com/addyosmani) (61,714 stars)\n- [**goose**](https://github.com/aaif-goose/goose) - Open-source extensible AI coding agent that goes beyond code suggestions. by [@aaif-goose](https://github.com/aaif-goose) (49,654 stars)\n- [**Aider**](https://github.com/Aider-AI/aider) - AI pair programming in your terminal. Edit code with LLMs across 100+ languages with deep Git integration. by [@Aider-AI](https://github.com/Aider-AI) (46,359 stars)\n- [**Taste-Skill**](https://github.com/Leonxlnx/taste-skill) - High-Agency frontend skill that helps AI generate less generic, more tasteful outputs. by [@Leonxlnx](https://github.com/Leonxlnx) (45,625 stars)\n- [**Qwen Code**](https://github.com/QwenLM/qwen-code) - Open-source AI coding agent that lives in your terminal, optimized for Qwen-Coder models. by [@QwenLM](https://github.com/QwenLM) (25,297 stars)\n- [**SWE-Agent**](https://github.com/SWE-agent/SWE-agent) - Automatically fix GitHub issues and handle cybersecurity challenges. State-of-the-art on SWE-bench. by [@SWE-agent](https://github.com/SWE-agent) (19,542 stars)\n- [**Devika**](https://github.com/stitionai/devika) - The first open-source implementation of an Agentic Software Engineer. An open-source alternative to Devin. by [@stitionai](https://github.com/stitionai) (19,513 stars)\n- [**context-mode**](https://github.com/mksglu/context-mode) - Context window optimization tool for AI coding agents with sandboxed tool output and 98% token reduction. by [@mksglu](https://github.com/mksglu) (17,634 stars)\n- [**Plandex**](https://github.com/plandex-ai/plandex) - Open-source AI coding agent designed for large projects and complex real-world tasks with persistent context. by [@plandex-ai](https://github.com/plandex-ai) (15,457 stars)\n- [**Trae Agent**](https://github.com/bytedance/trae-agent) - LLM-based agent by ByteDance for general-purpose software engineering tasks. by [@bytedance](https://github.com/bytedance) (11,672 stars)\n- [**Open SWE**](https://github.com/langchain-ai/open-swe) - Open-source asynchronous coding agent by LangChain for software engineering tasks. by [@langchain-ai](https://github.com/langchain-ai) (9,989 stars)\n- [**Mini-SWE-Agent**](https://github.com/SWE-agent/mini-swe-agent) - The 100-line AI agent that solves GitHub issues. Radically simple but scores \u003e74% on SWE-bench verified. by [@SWE-agent](https://github.com/SWE-agent) (5,237 stars)\n- [**Reflexion**](https://github.com/noahshinn/reflexion) - Language agents with verbal reinforcement learning. Agents that learn from mistakes through self-reflection. by [@noahshinn](https://github.com/noahshinn) (3,183 stars)\n\u003c!-- /AUTOGEN:coding --\u003e\n\n## Prompt and Behaviour Optimization\n\nTools and frameworks for automatically optimizing agent prompts, instructions, and behavioral patterns.\n\n\u003c!-- AUTOGEN:prompt-optimization --\u003e\n- [**Promptfoo**](https://github.com/promptfoo/promptfoo) - Open-source LLM evaluation and red-teaming framework. Test prompts, agents, and RAGs with 90+ model providers and 67+ security plugins. by [@promptfoo](https://github.com/promptfoo) (22,304 stars)\n- [**TextGrad**](https://github.com/zou-group/textgrad) - Automatic differentiation via text. Backpropagation through LLM-provided textual gradients, published in Nature. by [@zou-group](https://github.com/zou-group) (3,610 stars)\n\u003c!-- /AUTOGEN:prompt-optimization --\u003e\n\n## Agent Safety and Guardrails\n\nProjects focused on controlling agent actions, enforcing policies, and preventing harmful behavior.\n\n\u003c!-- AUTOGEN:safety --\u003e\n- [**NeMo Guardrails**](https://github.com/NVIDIA-NeMo/Guardrails) - NVIDIA's toolkit for adding programmable guardrails to LLM conversational systems. Policy-based safety controls. by [@NVIDIA-NeMo](https://github.com/NVIDIA-NeMo) (6,454 stars)\n- [**AgentDoG**](https://github.com/AI45Lab/AgentDoG) - Diagnostic guardrail framework for AI agent safety and security. Detects and intercepts unsafe agent behavior at runtime. by [@AI45Lab](https://github.com/AI45Lab) (626 stars)\n\u003c!-- /AUTOGEN:safety --\u003e\n\n## Embodied AI\n\nProjects connecting AI agents to physical devices, robotics, and real-world environments.\n\n\u003c!-- AUTOGEN:embodied --\u003e\n- [**Open-AutoGLM**](https://github.com/zai-org/Open-AutoGLM) - An Open Phone Agent Model and Framework. Unlocking the AI Phone for Everyone. by [@zai-org](https://github.com/zai-org) (25,550 stars)\n- [**LeRobot**](https://github.com/huggingface/lerobot) - Open-source robotics framework by Hugging Face. Models, datasets, and tools for real-world robotics in PyTorch. (25,053 stars)\n- [**Nanobrowser**](https://github.com/nanobrowser/nanobrowser) - Chrome extension for AI-powered web automation. Run multi-agent workflows using your own AI keys. by [@nanobrowser](https://github.com/nanobrowser) (13,307 stars)\n- [**XcodeBuildMCP**](https://github.com/getsentry/XcodeBuildMCP) - A MCP server and CLI for agent use when working on iOS and macOS projects. by [@getsentry](https://github.com/getsentry) (5,917 stars)\n- [**Mobile MCP**](https://github.com/mobile-next/mobile-mcp) - Model Context Protocol Server for Mobile Automation and Scraping (iOS, Android, Emulators and Real Devices). by [@mobile-next](https://github.com/mobile-next) (5,219 stars)\n- [**agent-device**](https://github.com/callstack/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](https://github.com/callstack) (2,807 stars)\n- [**ROS-LLM**](https://github.com/Auromix/ROS-LLM) - Framework for embodied intelligence in ROS. Natural language interactions with LLMs for robot control. by [@Auromix](https://github.com/Auromix) (802 stars)\n- [**RAI**](https://github.com/RobotecAI/rai) - Vendor-agnostic agentic framework for Physical AI and robotics. Connects LLM agents to ROS 2 tools for perception, reasoning, and control. by [@RobotecAI](https://github.com/RobotecAI) (529 stars)\n\u003c!-- /AUTOGEN:embodied --\u003e\n\n## Key Research Papers\n\n### Surveys\n\n- [A Comprehensive Survey of Self-Evolving AI Agents](https://arxiv.org/abs/2508.07407) (arXiv'25) - Unified framework with four components: System Inputs, Agent System, Environment, and Optimisers. Covers evolution of models, prompts, memory, tools, and workflows.\n- [A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve](https://arxiv.org/abs/2507.21046) (TMLR'26) - Organizes around what to evolve, when to evolve, and how to evolve. Covers intra-test-time and inter-test-time adaptation.\n- [Memory for Autonomous LLM Agents: Mechanisms, Evaluation, and Emerging Frontiers](https://arxiv.org/abs/2603.07670) (arXiv'26) - Formalizes agent memory as write-manage-read loop. Taxonomy spanning temporal scope, representational substrate, and control policy.\n\n### Self-Evolution and Lifelong Learning\n\n- [Live-SWE-agent: Can Software Engineering Agents Self-Evolve on the Fly?](https://arxiv.org/abs/2511.13646) (arXiv'25) - First live agent that autonomously evolves itself during runtime. 77.4% on SWE-bench Verified.\n- [EvoClaw: Evaluating AI Agents on Continuous Software Evolution](https://arxiv.org/abs/2603.13428) (arXiv'26) - Benchmark revealing performance drops from \u003e80% to at most 38% in continuous evolution settings.\n- [Symbolic Learning Enables Self-Evolving Agents](https://arxiv.org/abs/2406.18532) (arXiv'24) - Agents that evolve through symbolic representation learning.\n- [Building Self-Evolving Agents via Experience-Driven Lifelong Learning](https://arxiv.org/abs/2504.01072) (arXiv'25) - Framework and benchmark for lifelong agent learning.\n- [Darwin Godel Machine](https://arxiv.org/abs/2505.22954) (arXiv'25) - Agents that rewrite their own code through evolutionary pressure.\n- [EvoAgent: Self-evolving Agent with Continual World Model](https://arxiv.org/abs/2502.05907) (arXiv'25) - Continual world model for long-horizon task evolution.\n- [Absolute Zero: Reinforced Self-play Reasoning with Zero Data](https://arxiv.org/abs/2505.03335) (arXiv'25) - Self-play reasoning without any training data.\n- [AutoAgent: Evolving Cognition and Elastic Memory Orchestration](https://arxiv.org/abs/2603.09716) (arXiv'26) - Self-evolving framework with evolving cognition and elastic memory.\n- [Group-Evolving Agents](https://arxiv.org/abs/2602.04837) (arXiv'26) - Agent groups as evolutionary units with experience sharing. 71.0% on SWE-bench Verified.\n- [Agent0: Unleashing Self-Evolving Agents from Zero Data](https://arxiv.org/abs/2511.16043) (arXiv'25) - Curriculum and executor competition for self-improvement.\n- [SEMAG: Self-Evolutionary Multi-Agent Code Generation](https://arxiv.org/abs/2603.15707) (arXiv'26) - Self-evolutionary agents that auto-upgrade backbone models. 52.6% on CodeContests.\n- [SAGE: Multi-Agent Self-Evolution for LLM Reasoning](https://arxiv.org/abs/2603.15255) (arXiv'26) - Four co-evolving agents from shared LLM backbone.\n\n### Memory Optimization\n\n- [Agentic Memory: Unified Long-Term and Short-Term Memory Management](https://arxiv.org/abs/2601.01885) (arXiv'26) - Memory operations as tool-based actions with progressive RL training via GRPO.\n- [MEMORA: Harmonic Memory Representation](https://arxiv.org/abs/2602.03315) (arXiv'26) - Balances abstraction and specificity. SOTA on LoCoMo and LongMemEval.\n- [Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory](https://arxiv.org/abs/2504.19413) (arXiv'25) - Production architecture. 26% improvement on LOCOMO, 91% latency reduction.\n- [TeleMem: Long-Term and Multimodal Memory for Agentic AI](https://arxiv.org/abs/2601.06037) - Multimodal memory achieving 19% higher accuracy, 43% fewer tokens, 2.1x speedup over Mem0. (arXiv 2026)\n- [A-MEM: Agentic Memory for LLM Agents](https://arxiv.org/abs/2502.12110) (arXiv'25) - Self-organizing memory with autonomous management.\n- [Agent Workflow Memory](https://arxiv.org/abs/2409.07429) (ICML'24) - Memory tied to agent workflow patterns.\n- [MemoryBank: Enhancing Large Language Models with Long-Term Memory](https://arxiv.org/abs/2305.10250) (AAAI'24) - Structured long-term memory for LLMs.\n- [Compress to Impress](https://arxiv.org/abs/2402.11975) (ICLR'25) - Compression-based memory for extended dialogues.\n\n### Prompt and Behaviour Evolution\n\n- [ARTEMIS: Evolutionary Optimization for LLM Agent Configurations](https://arxiv.org/abs/2512.09108) (arXiv'25) - Semantically-aware genetic operators for joint agent config optimization. 13.6% on competitive programming.\n- [E-SPL: Unifying Evolutionary Prompt Search and Reinforcement Learning](https://arxiv.org/abs/2602.14697) (arXiv'26) - Joint RL weight updates with genetic operators for system prompt evolution.\n- [EvoPrompt: Connecting LLMs with Evolutionary Algorithms](https://arxiv.org/abs/2309.08532) (ICLR'24) - Evolutionary algorithms for prompt optimization.\n- [Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution](https://arxiv.org/abs/2309.16797) (ICML'24) - Prompts that evolve themselves recursively.\n- [Large Language Models as Optimizers (OPRO)](https://arxiv.org/abs/2309.03409) (ICLR'24) - Using LLMs to optimize their own prompts.\n- [TextGrad: Automatic Differentiation via Text](https://arxiv.org/abs/2406.07496) (Nature'25) - Gradient-like optimization through text feedback.\n\n### Tool and Code Evolution\n\n- [AlphaEvolve](https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/alphaevolve-a-gemini-powered-coding-agent-for-designing-advanced-algorithms/AlphaEvolve.pdf) (Google'25) - LLM-driven evolutionary code improvement.\n- [Learning Evolving Tools for Large Language Models](https://arxiv.org/abs/2410.06617) (ICLR'25) - Tools that co-evolve with agent capabilities.\n- [CREATOR: Tool Creation for Disentangling Abstract and Concrete Reasoning](https://arxiv.org/abs/2305.14318) (EMNLP'23) - Agents that create their own tools.\n- [ToolRL: Reward is All Tool Learning Needs](https://arxiv.org/abs/2504.13958) (arXiv'25) - Reinforcement learning for tool use optimization.\n\n### Reasoning and Planning\n\n- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://arxiv.org/abs/2303.11366) (NeurIPS'23) - Agents that learn from mistakes through self-reflection.\n- [ReflAct: World-Grounded Decision Making via Goal-State Reflection](https://arxiv.org/abs/2505.15182) (arXiv'25) - Goal-state reflection improving strategic reliability by 27.7% over ReAct.\n\n### Safety, Red-Teaming, and Alignment\n\n- [AgenticRed: Optimizing Agentic Systems for Automated Red-teaming](https://arxiv.org/abs/2601.13518) (arXiv'26) - Evolutionary red-teaming workflow design. 96% attack success on Llama-2-7B.\n- [Agent vs. Agent: Automated Red-Teaming for Custom Agentic Workflows](https://aclanthology.org/2025.emnlp-industry.62/) (EMNLP'25) - AgentHarm-Gen for adversarial task generation. 162% increase in attack success rate.\n- [AGENTSAFE: Unified Framework for Ethical Assurance and Governance](https://arxiv.org/abs/2512.03180) (arXiv'25) - Design, runtime, and audit controls covering the agentic loop.\n- [OpenGuardrails: Context-Aware AI Guardrails Platform](https://arxiv.org/abs/2510.19169) (arXiv'25) - Context-aware safety detection and model-manipulation prevention.\n\n### Embodied AI and Robotics\n\n- [RACAS: Controlling Diverse Robots With a Single Agentic System](https://arxiv.org/abs/2603.05621) (arXiv'26) - Single architecture controlling ground robots, robotic limbs, and underwater vehicles via natural language.\n- [RoboClaw: Scalable Long-Horizon Robotic Tasks](https://arxiv.org/abs/2603.11558) (arXiv'26) - VLM-driven framework with 25% improvement on long-horizon tasks and 53.7% less human time.\n- [MEM: Multi-Scale Embodied Memory for Vision Language Action Models](https://arxiv.org/abs/2603.03596) (arXiv'26) - Mixed-modal memory for tasks spanning up to fifteen minutes.\n\n## Benchmarks and Evaluation\n\n- [SWE-bench](https://github.com/princeton-nlp/SWE-bench) (ICLR'24) - Can agents resolve real-world GitHub issues?\n- [AgentBench](https://github.com/THUDM/AgentBench) (ICLR'24) - Multi-dimensional evaluation of LLMs as agents.\n- [WebArena](https://github.com/web-arena-x/webarena) (ICLR'24) - Realistic web environment for autonomous agents.\n- [OSWorld](https://github.com/xlang-ai/OSWorld) (NeurIPS'24) - Open-ended tasks in real computer environments.\n- [GAIA](https://huggingface.co/gaia-benchmark) (ICLR'23) - General AI assistant capabilities benchmark.\n- [EvoClaw](https://github.com/FSoft-AI4Code/EvoClaw) (arXiv'26) - Evaluating agents on continuous software evolution.\n- [LoCoMo](https://github.com/snap-research/locomo) (arXiv'25) - Long-context memory benchmark for agent memory systems.\n\n## Community and Knowledge\n\n\u003c!-- AUTOGEN:community --\u003e\n- [**Awesome-Self-Evolving-Agents**](https://github.com/EvoAgentX/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](https://github.com/EvoAgentX) (2,245 stars)\n\u003c!-- /AUTOGEN:community --\u003e\n\n## Footnotes\n\nMaintained by [EvoMap](https://evomap.ai). See [contributing guidelines](contributing.md) for how to submit a project or paper.\n\nAlso check out [Awesome Agent Swarm](https://github.com/EvoMap/awesome-agent-swarm) for multi-agent orchestration, swarm intelligence, and collaborative agent systems.\n\n[![Star History Chart](https://api.star-history.com/svg?repos=EvoMap/awesome-agent-evolution,EvoMap/awesome-agent-swarm,EvoMap/evolver,EvoMap/gep-mcp-server,mem0ai/mem0,letta-ai/letta,memvid/memvid,NevaMind-AI/memU,topoteretes/cognee,elizaOS/eliza,agent0ai/agent-zero,codelion/openevolve,facebookresearch/HyperAgents,anthropics/claude-code,openai/codex,google/A2A,mcp-use/mcp-use,All-Hands-AI/OpenHands,langchain-ai/langchain\u0026type=Date)](https://star-history.com/#EvoMap/awesome-agent-evolution\u0026EvoMap/awesome-agent-swarm\u0026EvoMap/evolver\u0026mem0ai/mem0\u0026letta-ai/letta\u0026memvid/memvid\u0026elizaOS/eliza\u0026agent0ai/agent-zero\u0026codelion/openevolve\u0026facebookresearch/HyperAgents\u0026anthropics/claude-code\u0026openai/codex\u0026google/A2A\u0026All-Hands-AI/OpenHands\u0026langchain-ai/langchain\u0026Date)\n","projects_url":"https://awesome.ecosyste.ms/api/v1/lists/evomap%2Fawesome-agent-evolution/projects"}