- Langchain★147,4 Tsd.
The agent engineering platform
- CrewAI★59,3 Tsd.
Framework for orchestrating role-playing, autonomous AI…
- Autogen★61,3 Tsd.
A programming framework for agentic AI
- Llama Index★52,4 Tsd.
LlamaIndex is the document processing platform for AI
- Langgraph★42,7 Tsd.
Build resilient agents
- Openai Agents Python★29,8 Tsd.
A lightweight, powerful framework for multi-agent…
- Eliza★19,5 Tsd.
Open source agentic operating system
- CAMEL★17,8 Tsd.
🐫 CAMEL: The first and the best multi-agent framework.…
- Pipecat★16,2 Tsd.
Open Source framework for voice agents, multimodal…
- Agents★14,5 Tsd.
A framework for building realtime voice AI agents 🤖🎙️📹
- Haystack★26,6 Tsd.
Open-source AI orchestration framework for building…
- superpowers★294,9 Tsd.
An agentic skills framework & software development…
- MetaGPT★70,7 Tsd.
🌟 The Multi-Agent Framework: First AI Software Company,…
- hyperframes★56,3 Tsd.
Write HTML. Render video. Built for agents.
- nanobot★48,8 Tsd.
Ultra-lightweight, open-source, self-hosted personal AI…
- AstrBot★41,3 Tsd.
AI Agent Assistant & development framework that…
- Anthropic-Cybersecurity-Skills★33,8 Tsd.
817 structured cybersecurity skills for AI agents ·…
- mastra★28,5 Tsd.
Mastra is the modern TypeScript framework for…
- TencentDB-Agent-Memory★27,7 Tsd.
TencentDB Agent Memory is a team-level memory hub for…
- Open-AutoGLM★26,3 Tsd.
An Open Phone Agent Model & Framework. Unlocking the AI…
- go-micro★23,1 Tsd.
A framework for building agents and services
- agent-zero★19,4 Tsd.
Agent Zero AI framework
- ai-berkshire★16,6 Tsd.
AI 时代的伯克希尔:基于 Claude Code / Codex…
- RagaAI-Catalyst★16,2 Tsd.
Python SDK for Agent AI Observability, Monitoring and…
- agent-framework★13,9 Tsd.
A framework for building, orchestrating and deploying…
- pest★11,7 Tsd.
The elegant testing framework for PHP developers and AI…
- ten-framework★11,1 Tsd.
Open-source framework for conversational voice AI agents
- spring-ai-alibaba★11,0 Tsd.
Agentic AI Framework for Java Developers
- voltagent★10,7 Tsd.
AI Agent Engineering Platform built on an Open Source…
- mcp-use★10,7 Tsd.
The fullstack MCP framework to develop MCP Apps for…
- omnigent★10,4 Tsd.
Omnigent is an open-source AI agent framework and…
- PraisonAI★9,1 Tsd.
PraisonAI 🦞 — Hire a 24/7 AI Workforce. Stop writing…
- agent-squad★7,8 Tsd.
Flexible and powerful framework for managing multiple…
- swarms★7,2 Tsd.
The Enterprise-Grade Multi-Agent Orchestration…
- Product-Manager-Skills★7,2 Tsd.
Product Management skills framework built on…
- agent-native★7,0 Tsd.
A framework for building agentic apps
- genkit★6,5 Tsd.
Open-source framework for building agentic apps in…
- agentops★5,9 Tsd.
Python SDK for AI agent monitoring, LLM cost tracking,…
- pyspur★5,8 Tsd.
A visual playground for agentic workflows: Iterate over…
- ruoyi-ai★5,7 Tsd.
Enterprise-grade AI agent framework with multi-provider…
- OpenAgentsControl★4,9 Tsd.
AI agent framework for plan-first development workflows…
- koog★4,6 Tsd.
Koog is a JVM (Java and Kotlin) framework for building…
- embabel-agent★4,5 Tsd.
Agent framework for the JVM. Pronounced Em-BAY-bel…
- ruby_llm★4,4 Tsd.
The Ruby-native AI framework. Chats, agents, tools,…
- AdalFlow★4,2 Tsd.
AdalFlow: The library to build & auto-optimize LLM…
- browser-agent★4,1 Tsd.
Open-source, vision-first browser agent
- TradingAgents-astock★3,6 Tsd.
A股多Agent投研框架 —…
- obsidian-wiki★3,5 Tsd.
Framework for AI agents to build and maintain a digital…
- beeai-framework★3,4 Tsd.
Build production-ready AI agents in both Python and…
- prompts.chat★171,9 Tsd.
f.k.a. Awesome ChatGPT Prompts. Share, discover, and…
- system-prompts-and-models-of-ai-tools★144,0 Tsd.
FULL Augment Code, Claude Code, Cluely, CodeBuddy,…
- awesome-llm-apps★140,6 Tsd.
100+ AI Agents, Agent Skills and RAG Apps - Free and…
- crawl4ai★84,7 Tsd.
Open-source web crawler and scraper for LLMs and AI…
- deer-flow★83,4 Tsd.
An open-source long-horizon SuperAgent harness that…
- AppFlowy★77,1 Tsd.
Bring projects, wikis, and teams together with AI.…
- career-ops★73,4 Tsd.
Open-source AI job search agent and job finder: scan…
- strix★66,3 Tsd.
Open-source AI penetration testing tool to find and fix…
- mempalace★59,4 Tsd.
The best-benchmarked open-source AI memory system. And…
- voicebox★56,3 Tsd.
The open-source AI voice studio. Clone, dictate, create.
- goose★54,9 Tsd.
an open source, extensible AI agent that goes beyond…
- VibeVoice★54,6 Tsd.
Open-Source Frontier Voice AI
- multica★51,9 Tsd.
Make humans and AI agents work as one team —…
- LocalAI★49,4 Tsd.
LocalAI is the open-source AI engine. Run any model -…
- siyuan★46,6 Tsd.
An open-source, privacy-first, self-hosted knowledge…
- reactive-resume★43,7 Tsd.
A one-of-a-kind resume builder that keeps your privacy…
- ToolJet★41,0 Tsd.
Open-source foundation of ToolJet AI - the enterprise…
- deepteam★3,0 Tsd.
Open-source red-teaming framework that stress-tests…
- zeron★3,0 Tsd.
Native control plane for Claude Code, Codex, Cursor,…
- opengap★3,0 Tsd.
Framework-agnostic, git-native standard for defining…
- ruler★2,9 Tsd.
Write coding-agent rules once, then apply them to every…
- clawpanel★2,9 Tsd.
Cross-platform desktop panel for managing OpenClaw and…
- rill★2,9 Tsd.
Blazing-fast business intelligence built for humans and…
- eventcatalog★2,9 Tsd.
Living architecture documentation that both your team…
- redamon★2,9 Tsd.
Self-hosted AI pentesting agent that finds flaws and…
- agent-rules-books★2,9 Tsd.
Battle-tested coding rules for AI agents, distilled…
- CloddsBot★2,9 Tsd.
Autonomous AI trading agent spanning 1000+ markets,…
- AI-fundamentals★2,9 Tsd.
Chinese-language course on AI basics: GPU architecture,…
- cindy★2,9 Tsd.
Out-of-the-box open-source AI agent for desktop and…
- qwen-audio-agent★2,8 Tsd.
Real-time voice runtime that keeps AI agents talking,…
- openlit★2,8 Tsd.
Open-source observability for AI agents: trace,…
- tons-of-skills-marketplace★2,8 Tsd.
Model-agnostic marketplace of agent skills with a…
- Paper2Any★2,8 Tsd.
Turn papers, text, or topics into editable research…
- tirith★2,8 Tsd.
Terminal security that stops malicious prompts, URLs,…
- claude-code-from-scratch★2,7 Tsd.
Rebuild Claude Code’s core in ~5,000 lines: an…
- yu-ai-agent★2,7 Tsd.
Chinese Java course: build AI agents with Spring AI,…
- memsearch★2,7 Tsd.
Persistent memory layer for AI agents, backed by…
- skillshare★2,7 Tsd.
Sync your AI coding setup: skills, agents, rules, MCP,…
- danghuangshang★2,7 Tsd.
Multi-agent court inspired by ancient China:…
- agent-skills★2,7 Tsd.
Official agent skills that teach AI coding tools to…
- yao-meta-skill★2,7 Tsd.
YAO: rigorous engineering, evaluation, and governance…
- LearnPrompt★2,7 Tsd.
Free Chinese AIGC course: prompting, agents, image,…
- rmux★2,7 Tsd.
Rust terminal multiplexer with a typed SDK: drive any…
- Memoh★2,6 Tsd.
Multi-agent platform where every agent gets its own…
- xerj★2,6 Tsd.
AI-native search that indexes code, docs, and PDFs with…
- Data-Analysis-Agent★2,6 Tsd.
Chat with your data: an LLM agent that builds charts…
- agentfield★2,6 Tsd.
Build, run, and scale AI agents like APIs and…
- agentic-context-engine★2,6 Tsd.
Make your agents learn from experience with a dedicated…
- griptape★2,6 Tsd.
Modular Python framework for AI agents and workflows…
- deepagents-in-action★2,6 Tsd.
Chinese guide to building production-grade AI agents…
- MiroFish-Offline★2,6 Tsd.
Offline multi-agent simulation and prediction engine:…
- burr★2,6 Tsd.
Apache framework for apps that make decisions: agents,…
- AingDesk★2,5 Tsd.
Simple desktop AI assistant with knowledge bases,…
- Seedance2-Storyboard-Generator★2,5 Tsd.
Open skill that turns novels and stories into…
- nanocoder★2,5 Tsd.
Community-built terminal coding agent: bring your own…
- webgradients★2,5 Tsd.
180 free gradients for CSS, Figma, and Sketch, plus a…
- stop-that-shit★2,5 Tsd.
Hook and skill guard that stops coding agents from…
- zeroboot★2,5 Tsd.
Sub-millisecond VM sandboxes for AI agents via…
- fulling★2,4 Tsd.
AI full-stack engineer agent built on Next.js, Claude,…
- workflow★2,4 Tsd.
Vercel’s SDK for durable, reliable, observable apps and…
- obsidian-agent-client★2,4 Tsd.
Bring Claude Code, Codex, and Gemini CLI into Obsidian…
- magentic★2,4 Tsd.
Use LLMs as plain Python functions: prompts become…
- openakita★2,0 Tsd.
Open-source AI assistant framework with skills and an…
- css★1,9 Tsd.
A markup-driven CSS orchestration language for…
- yomo★1,9 Tsd.
Serverless AI agent framework running on…
- jido★1,9 Tsd.
Autonomous agent framework for Elixir: distributed,…
- TuyaOpen★1,9 Tsd.
AI-plus-IoT framework bringing AI agents to ESP32 and…
- trpc-agent-go★1,8 Tsd.
Go framework for production agent systems: graphs,…
- OpenGUI★1,8 Tsd.
Android GUI agent framework: AI that sees, plans, and…
- AgentDock★1,8 Tsd.
Retired agent framework: its prompts and guides live on…
- Adala★1,6 Tsd.
Autonomous data-labeling agents: let AI do the…
- patchwork★1,6 Tsd.
Agentic AI framework for automating enterprise…
- AVA★1,6 Tsd.
AI-native visual analytics framework that helps agents…
- agmsg★1,5 Tsd.
Let Claude Code, Codex, Gemini, and Copilot message…
- lime★1,5 Tsd.
Full-stack AI agent for coding, research, content, and…
- KaibanJS★1,5 Tsd.
JavaScript framework for multi-agent systems with a…
- agents★1,4 Tsd.
Build AI agents in a no-code visual builder or…
- InternAgent★1,4 Tsd.
Agentic framework for long-horizon autonomous…
- bernstein★1,4 Tsd.
Governance and orchestration for AI agent fleets:…
- llm-wiki★1,4 Tsd.
Parallel multi-agent research that compiles wikis from…
- cli-agent-orchestrator★1,4 Tsd.
Orchestrate Claude Code, Kiro, and Codex together in…
- 10x★1,4 Tsd.
Up to 20x faster AI coding with smart model routing and…
- nodeterm★1,9 Tsd.
Manage parallel AI coding agents on an infinite…
- adk-java★1,7 Tsd.
A code-first Java toolkit for building AI agents.
- agentic_coding_flywheel_setup★1,7 Tsd.
Turn a fresh Ubuntu VPS into a multi-agent dev…
- neo-chat★1,9 Tsd.
Local-first AI chat workspace for models.
- dmux★1,8 Tsd.
Dev agent multiplexer designed for git worktrees and…
- zerostack★1,7 Tsd.
Lean coding agent written in Rust, optimized for memory…
- claude-code-settings★1,7 Tsd.
Curated skills, sub-agents, and config templates that…
- duckle★1,4 Tsd.
Open source ETL/ELT you deploy on your own servers or…
- vellum-assistant★1,4 Tsd.
AI Assistant that’s easy to setup.
- nullhub★1,8 Tsd.
Management console for installing and monitoring AI…
- agent-beacon★1,8 Tsd.
Self-improving memory layer that works across AI agent…
- zero★1,7 Tsd.
A coding agent that runs on your model, your machine,…
- LongHorizon-Harness★1,7 Tsd.
Run AI agents on desktop apps and CLI for hours without…
- matlab-mcp-server★1,6 Tsd.
Official MathWorks MCP server: drive MATLAB from AI…
- ktx★1,6 Tsd.
Context layer so AI agents query your analytics…
- MemoryOS★1,6 Tsd.
A memory operating system for long-term, personalized…
- cc-safety-net★1,6 Tsd.
Pre-execution guard that stops AI agents' destructive…
- rocketmq-rust★1,5 Tsd.
Apache RocketMQ rewritten in Rust for the AI era.
- superlog★1,5 Tsd.
Observability that dispatches AI agents to self-heal…
- paritok-4b-v1★1,5 Tsd.
Compression gateway that cuts AI coding agents' token…
- openserp★1,4 Tsd.
Self-hosted SERP API: browser-rendered search from six…
- adk-js★1,4 Tsd.
Google's code-first TypeScript toolkit for building AI…
- commonly★1,4 Tsd.
Shared rooms where humans and cross-vendor AI agents…
- c-code-style★1,4 Tsd.
C99+ style guide with an AI agent skill for formatting…
- nezha★1,9 Tsd.
Code Editor for the AI Agents Era.
- mex★1,8 Tsd.
Team memory for engineers and their AI agents.
- CyberVerse★1,7 Tsd.
Self hosted, live digital human agent system.
- codedb★1,4 Tsd.
Zig code intelligence server and MCP toolset for AI…
- opendia★1,9 Tsd.
Connect your browser to AI models with a Dia-style…
- iwe★1,7 Tsd.
A markdown knowledge graph with LSP for editors and MCP…
- Swift-Concurrency-Agent-Skill★1,7 Tsd.
Give your AI coding tool expert Swift Concurrency…
- chops★1,9 Tsd.
Run chops locally: your ai agent skills, finally…
- gortex★1,9 Tsd.
Gortex, in Go — built for high-performance…
- CoreCoder★1,8 Tsd.
Self-hosted ai agents: minimal ai coding agent (~1,000…
- context-engineering-kit★1,7 Tsd.
Run context-engineering-kit locally: hand-crafted…
- spec-kitty★1,7 Tsd.
Open ai agents for spec-driven development with…
- tradememory-protocol★1,4 Tsd.
Tradememory-protocol — decision audit trail +…
- nocturne_memory★1,4 Tsd.
Nocturne_memory, in Python — built for a lightweight,…
- poco-claw★1,4 Tsd.
Run poco-claw locally: a more beautiful and…
- oh-my-agent★1,3 Tsd.
Open ai agents for mechanical verification for ai…
- openclaw★391,6 Tsd.
A personal AI that really does things — on any OS, with…
- firecrawl★189,6 Tsd.
Turn any website into clean, agent-ready data with one…
- pi★113,3 Tsd.
A TypeScript toolkit for building agents: unified LLM…
- openinterpreter★68,5 Tsd.
A coding agent tuned for open models like Kimi, GLM,…
- hindsight★47,0 Tsd.
Agent memory that learns — so your agents get smarter…
- agno★42,6 Tsd.
Build, run, and manage full agent platforms — not just…
- dspy★38,6 Tsd.
Program language models with code and compilers — not…
- QwenPaw★35,5 Tsd.
Your personal AI assistant — self-host it, chat…
- graphiti★31,5 Tsd.
Real-time knowledge graphs that give agents a memory…
- deepagents★30,0 Tsd.
The batteries-included agent harness from the LangChain…
- semantic-kernel★28,6 Tsd.
Microsoft's SDK for weaving LLMs into real…
- A2A★26,1 Tsd.
The open protocol that lets AI agents talk to each…
- RAG-Anything★23,5 Tsd.
One RAG framework for everything — including multimodal…
- rasa★21,3 Tsd.
The veteran open-source framework for conversational AI…
- openwiki★17,0 Tsd.
A CLI that writes and maintains agent-ready docs for…
- unstructured★15,5 Tsd.
Turn messy documents into clean, structured data for…
- phoenix★11,7 Tsd.
Observability and evaluation for AI apps — see what…
- bifrost★8,6 Tsd.
A blazing-fast enterprise AI gateway: 1000+ models,…
- yao★8,1 Tsd.
All your agents and workspaces in one place — on every…
- guardrails★7,5 Tsd.
Put guardrails on LLMs: validate outputs before they…
- x402★6,7 Tsd.
A payments protocol for the internet — built on plain…
- OpenBot★6,2 Tsd.
Open-source AI coworkers, each with a computer of their…
- eve★5,5 Tsd.
Vercel's open framework for building agents that run…
- zep★5,0 Tsd.
Examples and integrations for Zep — the memory layer…
- cumora★3,9 Tsd.
Team chat where AI agents are first-class teammates,…
- vocode-core★3,8 Tsd.
Build voice-based LLM agents with a modular,…
- ReMe★3,6 Tsd.
A memory management kit for agents — remember me,…
- agentkit★1,3 Tsd.
Every AI agent deserves a wallet — onchain tools from…
- Agent_Memory_Techniques★1,1 Tsd.
30 runnable notebooks teaching everything about agent…
- arcade-mcp★1,0 Tsd.
A Python framework for building MCP servers and agent…
- mcp-gateway-registry★964
An enterprise MCP gateway: governed, auditable tool…
- agent-qa★902
A self-improving QA agent that writes tests in plain…
- modelscan★781
Scan AI models for serialization attacks before you…
- fastagency★547
Take multi-agent AutoGen workflows to production, fast.
- vigil-llm★504
Detect prompt injections and jailbreaks before they hit…
- MMLU-Pro★429
The tougher, more robust successor to the MMLU…
- OSWorld-V2★353
Benchmark computer-use agents on long, real-world…
- agent-memory-benchmark★144
A benchmark for comparing AI agent memory systems…
- ActPlane★104
eBPF-powered information-flow enforcement for safer AI…
- taosmd★79
Local-first AI memory that runs offline on just 8 GB of…
- kuberna-labs★51
Build, deploy, and monetize autonomous agents across…
- hermes-agent★252,0 Tsd.
The agent that grows with you — a Python framework for…
- ponytail★157,9 Tsd.
Skills and plugins that teach coding agents to write…
- claude-mem★97,9 Tsd.
Give your coding agents a persistent memory that…
- docling★68,5 Tsd.
Convert PDFs, DOCX, and HTML into clean, AI-ready…
- milvus★46,3 Tsd.
The cloud-native vector database for planet-scale…
- LightRAG★40,0 Tsd.
Simple, fast retrieval-augmented generation with graph…
- firecracker★37,2 Tsd.
Secure, fast microVMs that make serverless possible.
- qdrant★35,0 Tsd.
A high-performance vector search engine for next-gen AI.
- supermemory★31,2 Tsd.
The memory API that gives AI apps instant, lasting…
- chroma★29,5 Tsd.
Open-source search infrastructure built for AI…
- label-studio★28,4 Tsd.
The open-source studio for labeling data of every type.
- letta★25,1 Tsd.
Stateful AI agents with memory that learns and improves.
- adk-python★21,7 Tsd.
Google's code-first toolkit for building real AI agents.
- deepeval★18,7 Tsd.
Unit tests for LLMs — evaluate before you ship.
- weaviate★16,9 Tsd.
Vector search plus structured filtering in one database.
- E2B★14,2 Tsd.
Secure sandboxes where AI agents run real code.
- garak★9,5 Tsd.
Scan your LLM for vulnerabilities before attackers do.
- microsandbox★8,6 Tsd.
Local-first microVMs for fast, programmable sandboxes.
- R2R★8,0 Tsd.
Production-ready agentic RAG behind a clean REST API.
- openllmetry★7,5 Tsd.
OpenTelemetry-native observability for LLM apps.
- agent-governance-toolkit★6,4 Tsd.
Policy, identity, and sandboxing guardrails for AI…
- SWE-bench★6,0 Tsd.
The benchmark that grades AI on real GitHub issues.
- 01★5,2 Tsd.
Talk to your computer — an open-source voice interface.
- agenta★4,8 Tsd.
A team workspace for building AI agents together.
- notfair-plugin★3,9 Tsd.
SEO, GEO, and marketing skills your agents can use.
- AgentBench★3,8 Tsd.
A rigorous benchmark for LLMs acting as agents.
- AP2★3,2 Tsd.
An open protocol for secure AI-driven payments.
- labs-molt★1,2 Tsd.
A lean, HuggingFace-native RL framework for agentic…
- langsmith-sdk★1,1 Tsd.
Official SDKs for tracing with LangSmith.
- tada★1,0 Tsd.
An open-source speech language model from Hume.
- motorhead★917
A memory and retrieval server built for LLMs.
- agentdojo★897
Test your agents against prompt-injection attacks.
- agentsight★728
See what your AI agents do — at the system level.
- gpqa★540
A graduate-level benchmark even Google can't crack.
- invariant★467
Guardrails for building secure, robust agents.
- RAMPART★419
Safety testing for agents, right in pytest.
- evals★230
A complete evaluation toolkit for AI agents.
- OpenYabby★130
A voice-driven multi-agent assistant that orchestrates…
- AgentArk★92
A personal AI assistant that puts security first.
- Agon★54
Autonomous research loops, from topic to experiments.
- deepseek-harness★245,5 Tsd.
DeepSeek's plugin-based agent harness where everything…
- browser-use★117,4 Tsd.
The open-source library that lets AI agents control a…
- servers★91,1 Tsd.
The official collection of reference Model Context…
- mem0★66,8 Tsd.
Drop-in persistent memory infrastructure for AI agents…
- agent-browser★43,6 Tsd.
A fast browser-automation CLI built for AI agents.
- OpenViking★39,4 Tsd.
A self-evolving context database unifying agent memory,…
- sglang★36,9 Tsd.
High-performance serving for the LLMs that power your…
- cognee★31,6 Tsd.
Open-source AI memory that gives agents persistent…
- composio★30,5 Tsd.
Over a thousand tool integrations so agents can turn…
- mlx★28,7 Tsd.
Apple's array framework for running ML efficiently on…
- fastmcp★28,0 Tsd.
The fast, Pythonic way to build MCP servers and clients.
- python-sdk★24,5 Tsd.
The official Python SDK for Model Context Protocol…
- coze-studio★21,7 Tsd.
A visual all-in-one platform for building, debugging,…
- SuperAGI★17,7 Tsd.
A dev-first framework for building and running…
- OpenSandbox★15,7 Tsd.
A secure, fast sandbox runtime where AI agents can…
- typescript-sdk★13,5 Tsd.
The official TypeScript SDK for Model Context Protocol…
- harness-sdk★8,7 Tsd.
An open-source SDK for production agents on any model,…
- mcp-agent★8,6 Tsd.
Simple workflow patterns for building effective agents…
- steel-browser★7,8 Tsd.
A batteries-included browser API so agents can automate…
- Guardrails★7,3 Tsd.
Programmable guardrails to keep LLM systems safe and on…
- helicone★6,2 Tsd.
Open-source LLM observability: monitor, evaluate, and…
- harbor★5,9 Tsd.
A framework for evaluating and improving AI agents.
- ag2★5,0 Tsd.
The open-source AgentOS, formerly known as AutoGen.
- PyRIT★4,6 Tsd.
Microsoft's open framework for red-teaming generative…
- SimpleMem★3,8 Tsd.
Research-backed lifelong memory for LLM agents, text…
- morphik-core★3,7 Tsd.
An open-source multimodal retrieval engine for agent…
- labs-OO-Agents★2,3 Tsd.
NVIDIA's object-oriented, Pythonic approach to building…
- weave★1,1 Tsd.
A toolkit for developing AI-powered apps, from the…
- autoevals★1,1 Tsd.
Fast, best-practice evaluations for AI model outputs.
- ClawBench★975
An open benchmark for browser AI agents on everyday…
- macos-harness★902
A minimal harness that lets an LLM fully control a Mac.
- bolna★786
Open-source conversational voice AI agents.
- awesome-ai-tools★592
A very large curated list of AI tools, updated for 2026.
- SWE-bench_Pro-os★539
A benchmark testing whether agents can solve…
- Hypha★447
A harness-oriented framework for production-grade agent…
- A-mem-sys★401
Agentic memory for LLM agents, from a published…
- octomind★151
An MCP-native AI coding agent and runtime in a single…
- babyagi_archive★110
A frozen snapshot of the original BabyAGI…
- NarraNexus★88
A framework for agent nexuses where intelligence…
- server-sdk-python★53
The official Python SDK for building on Vapi's voice AI…
- stagehand★25,6 Tsd.
The SDK for extracting data and interacting with any…
- mcp★789
An MCP server that puts Hyperbrowser's browsers behind…
- khoj★37,6 Tsd.
Your self-hostable AI second brain for answers, agents,…
- playwright-mcp★37,9 Tsd.
Microsoft's Playwright-powered MCP server for browser…
- tau2-bench★2,2 Tsd.
A benchmark for tool-agent-user interaction in…
- ECC★275,2 Tsd.
Agent harness optimization system with skills, memory…
- AutoGPT★187,5 Tsd.
The open-source platform for building and running…
- TradingAgents★110,3 Tsd.
Multi-agent LLM framework that mirrors a trading firm:…
- Agent-Reach★94,0 Tsd.
Give your AI agent eyes on the whole internet: Twitter,…
- OpenHands★90,3 Tsd.
Open-source AI-driven development platform: autonomous…
- ruflo★74,1 Tsd.
Agent harness for multi-agent swarms: deploy,…
- chrome-devtools-mcp★53,1 Tsd.
Chrome DevTools for coding agents: an MCP server that…
- codebase-memory-mcp★46,1 Tsd.
High-performance code intelligence MCP server: your…
- openhuman★41,7 Tsd.
A fast, local-first open-source agent harness written…
- UI-TARS-desktop★39,2 Tsd.
ByteDance's open-source multimodal AI agent stack: GUI…
- openclaude★33,7 Tsd.
A Claude Code-style coding agent that runs anywhere and…
- github-mcp-server★33,4 Tsd.
GitHub's official MCP server — let AI tools work with…
- agentscope★32,9 Tsd.
Build and run agents you can see, understand and trust…
- nanoclaw★30,9 Tsd.
A lightweight, container-isolated personal AI agent for…
- genesis-world★30,0 Tsd.
A Python simulation platform for robotics and embodied…
- gpt-researcher★30,0 Tsd.
An autonomous research agent that plans, searches and…
- page-agent★29,4 Tsd.
Alibaba's in-page GUI agent: control any web interface…
- lerobot★28,0 Tsd.
Hugging Face's LeRobot: datasets, models and tools to…
- agenticSeek★27,5 Tsd.
A fully local Manus-style autonomous agent: browses,…
- promptfoo★25,8 Tsd.
Test prompts, agents and RAGs — LLM evals, red teaming…
- pentagi★25,3 Tsd.
Autonomous AI agents for penetration testing —…
- n8n-mcp★23,0 Tsd.
An MCP server that lets Claude, Cursor and Windsurf…
- NemoClaw★22,7 Tsd.
NVIDIA's reference stack for running AI agents more…
- opik★22,5 Tsd.
Open-source LLM observability: trace, evaluate and…
- babyagi★22,4 Tsd.
Yohei Nakajima's autonomous task agent - now a…
- generative_agents★22,2 Tsd.
Stanford's Generative Agents simulation - 25 AI…
- prime-agent★21,6 Tsd.
A self-improving agent for coding workflows and…
- SWE-agent★20,5 Tsd.
Give it a GitHub issue and it tries to fix the code for…
- suna★20,3 Tsd.
Kortix - the open-source AI operating system: agents,…
- Tongyi Deep Research★20,0 Tsd.
Tongyi Deep Research - Alibaba's open-source deep…
- SkillSpector★19,7 Tsd.
NVIDIA's security scanner for AI agent skills - check…
- devika★19,6 Tsd.
Devika - the open-source agentic software engineer that…
- agent-lightning★18,6 Tsd.
Microsoft's trainer for AI agents - reinforcement…
- browser-harness★18,4 Tsd.
A self-healing harness that connects an LLM directly to…
- parlant★18,3 Tsd.
Build customer-facing AI agents that actually follow…
- langchainjs★18,2 Tsd.
The JavaScript/TypeScript version of LangChain - the…
- openfang★18,2 Tsd.
An open-source agent operating system in Rust -…
- LangBot★18,1 Tsd.
A production-grade platform for building agentic chat…
- Memori★17,1 Tsd.
Agent-native memory infrastructure - structured,…
- mcp-toolbox★16,6 Tsd.
Google's MCP Toolbox for Databases - give agents safe,…
- memvid★16,6 Tsd.
A serverless, single-file memory layer for AI agents -…
- ragas★16,0 Tsd.
Supercharge LLM app evaluations with objective metrics,…
- PentestGPT★15,8 Tsd.
Automated penetration-testing agent framework powered…
- botpress★14,9 Tsd.
Open-source hub of integrations, SDK and CLI for…
- RD-Agent★14,9 Tsd.
Microsoft's R&D-Agent: LLM agents that automate…
- unity-mcp★14,8 Tsd.
Bridge AI assistants to the Unity Editor via MCP:…
- openpi★14,1 Tsd.
Physical Intelligence's open vision-language-action…
- BrowserOS★13,8 Tsd.
Open-source agentic browser your AI agents drive with…
- moli★13,7 Tsd.
Production-ready headless browser for AI agents, built…
- EverOS★13,4 Tsd.
One portable, local-first memory layer for every AI…
- ida-pro-mcp★12,5 Tsd.
AI-powered reverse engineering assistant that bridges…
- hexstrike-ai★12,5 Tsd.
MCP server letting AI agents run 150+ cybersecurity…
- bisheng★12,0 Tsd.
Enterprise LLM app platform: GenAI workflows, RAG,…
- MemOS★11,8 Tsd.
Self-evolving memory OS for LLMs and agents: persistent…
- holaOS★11,5 Tsd.
Local-first agentic workspace where Claude Code, Codex…
- apify-mcp-server★10,1 Tsd.
MCP server that lets AI agents run thousands of Apify…
- UFO★10,0 Tsd.
Microsoft's GUI agent that grew into UFO³ —…
- mobilerun★9,6 Tsd.
Control Android and iOS devices with natural language —…
- nasiko★9,5 Tsd.
The open runtime for AI agents — a Rust platform for…
- MobileAgent★9,3 Tsd.
Mobile-Agent — multimodal GUI agents that operate…
- local-deep-research★9,2 Tsd.
Deep research that stays on your machine — local or…
- evolver★9,1 Tsd.
A GEP-powered self-evolving engine for AI agents —…
- astron-agent★8,9 Tsd.
iFLYTEK's enterprise agentic workflow platform —…
- lamda★8,6 Tsd.
FIRERPA / LAMDA — full-stack Android device control:…
- IsaacLab★8,3 Tsd.
Isaac Lab — NVIDIA's unified framework for robot…
- Isaac-GR00T★8,2 Tsd.
NVIDIA Isaac GR00T N1.7 — an open foundation model for…
- Vision-Agents★8,2 Tsd.
Stream's open Vision Agents — build real-time voice and…
- osaurus★8,0 Tsd.
Osaurus — own your AI: a native macOS agent harness…
- Upsonic★8,0 Tsd.
Upsonic — build autonomous AI agents in Python, with…
- honcho★7,5 Tsd.
Honcho — Plastic Labs' memory library for stateful…
- reef★7,5 Tsd.
Reef — infrastructure for continually self-improving…
- Voyager★7,2 Tsd.
Voyager — the open-ended embodied Minecraft agent that…
- openvla★7,1 Tsd.
Open-source 7B vision-language-action model and…
- open-multi-agent★7,0 Tsd.
Open Multi Agent — a self-hosted TypeScript agent…
- AppAgent★6,9 Tsd.
AppAgent — Tencent's multimodal agent that learns…
- smolvm★6,6 Tsd.
An embeddable, portable, branchable microVM that runs…
- julep★6,6 Tsd.
Durable, composable AI agents — flows that crash and…
- MobileBuildMCP★6,5 Tsd.
An MCP server and CLI that lets AI agents build, test…
- LaVague★6,4 Tsd.
A Large Action Model framework for building AI web…
- Kun★6,3 Tsd.
A local-first AI agent workspace for coding, writing,…
- loopx★6,2 Tsd.
A control plane with a durable state kernel that keeps…
- ouroboros★6,2 Tsd.
An Agent OS with interview-gated, budgeted evolution —…
- LobsterAI★6,1 Tsd.
A desktop-grade AI agent from NetEase Youdao that does…
- open-connector★6,0 Tsd.
An open-source auth gateway connecting 1,500+ SaaS…
- nexent★5,9 Tsd.
A zero-code platform for auto-generating…
- klavis★5,8 Tsd.
An MCP integration platform that lets AI agents use…
- coze-loop★5,8 Tsd.
A full-lifecycle platform for developing, debugging,…
- TaskingAI★5,4 Tsd.
An open-source platform for AI-native application…
- Raven★5,3 Tsd.
The harness of harnesses: a persistent, self-evolving…
- zotero-mcp★5,3 Tsd.
Connect your Zotero research library to Claude and…
- FrontierAgent★5,2 Tsd.
An open agent framework with a native terminal TUI,…
- exa-mcp-server★5,1 Tsd.
Exa's MCP server for web search and crawling — give…
- ha-mcp★5,0 Tsd.
The unofficial Home Assistant MCP server — control and…
- eko★5,0 Tsd.
Build production-ready agentic workflows from natural…
- agent-device★4,9 Tsd.
Mobile app automation and verification built for AI…
- ghidra-mcp★4,9 Tsd.
Ghidra MCP server with 200+ tools that let AI…
- rivet★4,7 Tsd.
Rivet is Ironclad's visual IDE and TypeScript library…
- deep-research★4,7 Tsd.
Deep Research with any LLM: in-depth reports in…
- council-of-high-intelligence★4,6 Tsd.
Convene 18 legendary thinkers as an AI council that…
- LongMemory★4,5 Tsd.
LongMemory: local-first memory for AI agents with…
- m_flow★4,5 Tsd.
M-flow is a bio-inspired cognitive memory engine for AI…
- logfire★4,5 Tsd.
Pydantic Logfire: Python and AI observability on…
- memory-lancedb-pro★4,5 Tsd.
Memory-LanceDB-Pro gives OpenClaw agents hybrid search,…
- nono★4,4 Tsd.
nono: zero-trust, zero-setup, zero-latency sandboxing…
- archestra★4,4 Tsd.
Archestra: the enterprise AI platform — chat UI, LLM…
- arcbox★4,2 Tsd.
ArcBox: a fast macOS-native container and VM runtime —…
- excel-mcp-server★4,2 Tsd.
Excel MCP Server: let AI agents read, write, format,…
- basic-memory★4,1 Tsd.
Basic Memory: local-first MCP memory so you and your AI…
- langroid★4,1 Tsd.
Langroid: a principled Python framework for building…
- playwriter★4,0 Tsd.
Playwriter lets AI agents drive your real Chrome — your…
- fast-agent★3,9 Tsd.
Code, build and evaluate LLM agents with full MCP,…
- openai-agents-js★3,9 Tsd.
OpenAI's lightweight TypeScript framework for…
- LazyLLM★3,9 Tsd.
A low-code tool for building multi-agent LLM apps —…
- atmosphere★3,8 Tsd.
A portable AI agent runtime for the JVM — write one…
- antigravity-sdk-python★3,7 Tsd.
Google's Python SDK for building AI agents on the…
- goclaw★3,6 Tsd.
OpenClaw rebuilt in Go — multi-tenant AI agent teams…
- dbhub★3,6 Tsd.
A token-conscious MCP server that connects AI tools to…
- onecli★3,6 Tsd.
An open-source sandboxed agent harness for teams — a…
- EvoAgentX★3,4 Tsd.
EvoAgentX — a framework for building self-evolving…
- google_workspace_mcp★3,3 Tsd.
Control Gmail, Calendar, Docs, Sheets, Drive and more…
- mobile-use★3,2 Tsd.
AI agents that use real Android and iOS apps like a…
- arxiv-mcp-server★3,2 Tsd.
A local MCP server for serious arXiv work — LaTeX…
- OSWorld★3,2 Tsd.
OSWorld — benchmark multimodal agents on open-ended…
- agent-scan★3,1 Tsd.
Snyk's security scanner for AI agents, MCP servers and…
- Auto-Company★3,1 Tsd.
An auto-company that works 24/7 on your own PC — 14…
- EnvoyMesh★3,1 Tsd.
Decentralized P2P mesh for autonomous AI agents — your…
- memobase★2,9 Tsd.
User-profile-based long-term memory for AI chatbot…
- forkd★2,9 Tsd.
Fork() for AI agent microVMs — spawn 100 KVM-isolated…
- moltis★2,9 Tsd.
Secure persistent personal agent server in one Rust…
- jadx-ai-mcp★2,9 Tsd.
JADX plugin + MCP server that lets LLMs analyze Android…
- godot-ai★2,9 Tsd.
Production-grade MCP server for Godot — let AI build…
- PhyAgentOS-core★2,8 Tsd.
Recursive self-improving physical agent OS — verified,…
- paper-search-mcp★2,8 Tsd.
MCP server, CLI and skills for searching and…
- metamcp★2,7 Tsd.
MCP aggregator, orchestrator, middleware and gateway in…
- OmAgent★2,7 Tsd.
Python library for building multimodal language agents…
- webcmd★2,7 Tsd.
Self-learning browser infrastructure for AI agents —…
- NeMo-Agent-Toolkit★2,7 Tsd.
NVIDIA's toolkit for connecting, profiling and…
- brightdata-mcp★2,7 Tsd.
Bright Data's MCP server: web search, scraping,…
- Memary★2,7 Tsd.
The open-source memory layer for autonomous agents —…
- agent-memory★2,6 Tsd.
Long-term memory runtime for AI agents — plain Markdown…
- mcphub★2,5 Tsd.
Self-hosted MCP gateway and control plane — connect,…
- x64dbg-mcp-server★2,4 Tsd.
Native MCP plugin for x64dbg — let an AI drive the…
- pinvou-agent★2,4 Tsd.
Pinvou Agent — open-source desktop AI agent for tools,…
- AiSOC★2,4 Tsd.
Open-source AI SOC — alert fusion, LLM triage, MITRE…
- mantis★2,4 Tsd.
Google's modular toolkit that lets AI coding agents…
- open-codex-computer-use★2,4 Tsd.
Open-source Computer Use as an MCP service: let any…
- agent-vault★2,3 Tsd.
Infisical's credential vault and HTTP proxy, so AI…
- memanto★2,3 Tsd.
A memory agent for your agent fleet: it curates,…
- magic-context★2,3 Tsd.
Self-managing, unbounded context and long-term memory…
- dash★2,3 Tsd.
Agno's self-learning data agent: grounded text-to-SQL…
- Unla★2,2 Tsd.
Unla MCP Gateway: turn existing APIs and MCP servers…
- open-kritt★2,2 Tsd.
Self-hosted AI vulnerability research: agents find,…
- open-cowork★2,2 Tsd.
Open Cowork: a desktop AI agent app with one-click…
- mcp-shrimp-task-manager★2,1 Tsd.
Shrimp Task Manager: an MCP server that turns natural…
- future-agi★2,1 Tsd.
One open platform to simulate, evaluate, guardrail,…
- ADR★1,9 Tsd.
Uber's ADR: discover, observe, benchmark, and detect…
- gini-agent★1,9 Tsd.
Gini: a personal agent runtime that remembers, learns…
- babysitter★1,8 Tsd.
Babysitter enforces the process: deterministic,…
- agent-desktop★1,8 Tsd.
Reliable computer use for any agent: drive real apps…
- langmem★1,7 Tsd.
LangMem: long-term memory primitives for LangGraph…
- brave-search-mcp-server★1,5 Tsd.
Brave's official MCP server: web, news, image, video,…
- SandboxFusion★1,1 Tsd.
ByteDance's SandboxFusion: a secure sandbox for running…
- Qwen-AgentWorld★1,0 Tsd.
Qwen-AgentWorld: a language world model simulating MCP,…
- SimWorld★793
SimWorld: an Unreal Engine 5 simulator for evaluating…
- SWE-Gym★748
SWE-Gym: the first training environment for real-world…
- agent-world-model★465
Snowflake's Agent World Model: synthesise 1,000…
- WorkArena★275
WorkArena: ServiceNow's benchmark for how well web…
- webarena-verified★62
WebArena-Verified: the audited, deterministically…
- ZCode★7,9 Tsd.
ZCode — Z.ai's AI coding workspace: desktop app,…
- OpenScience★4,0 Tsd.
OpenScience — the open-source AI workbench for…
- Rakazo★4,1 Tsd.
Rakazo — an open-source, self-hosted AI agent for…
- Monid★4,2 Tsd.
Monid — OpenRouter for agent tools: one base URL and…
- Loop Engineering★11,6 Tsd.
Loop Engineering — practical patterns, starters and a…
- Talorys★1,1 Tsd.
Talorys — your personal AI agent in your own Cloudflare…
- Tutti★3,8 Tsd.
Tutti — a real-time shared workspace where people and…
- ESP32 iPhone Agent★2,4 Tsd.
ESP32 iPhone Agent — let an AI model use a real iPhone…
- OpenMausBot★4,9 Tsd.
OpenMausBot — the open-source Grok Bot alternative:…
- Beam★6
Beam — Reflection's open 501B MoE model built for…
- smolagents★29,9 Tsd.
smolagents — Hugging Face's barebones agent library…
- Jev Ultrafast★23,7 Tsd.
Jev Ultrafast — the fastest browser agent from…
- Swarm★22,1 Tsd.
Swarm — OpenAI's experimental framework for lightweight…
- ChatDev★34,6 Tsd.
ChatDev — zero-code multi-agent platform (DevAll) for…
- Anthropic Agent Skills★190,4 Tsd.
Anthropic's official Agent Skills for Claude —…
- Google Agent Skills★23,7 Tsd.
Google’s official Agent Skills for Google Cloud, Gemini…
- Security Audit Skill★67,6 Tsd.
Cloudflare's security-audit skill — turns your coding…
- GBrain★31,9 Tsd.
Garry Tan's agent memory — a self-hosted brain your…
- Auto Ontology★12,1 Tsd.
NVIDIA's context layer for agentic analytics —…
- Skills For Real Engineers★322,7 Tsd.
Matt Pocock's engineering skills for coding agents —…
- billion-context★7,0 Tsd.
Context compression for long coding sessions — 5x fewer…
- magpie★35,3 Tsd.
Every agent's model, one place — run Claude Code on…
- T3 Code★61,1 Tsd.
One fast interface for every coding agent on your…
- Vercel Agent Desktop★24,1 Tsd.
Computer use for AI agents through the OS accessibility…
- Claude Code Game Studios★39,5 Tsd.
Describe a game — Claude Code asks a few questions,…
- CLI Anything★58,8 Tsd.
Make any software agent-native: generate a CLI so…
- MiroFish★85,7 Tsd.
A swarm-intelligence prediction engine: thousands of…
- Mika★8,7 Tsd.
An AI assistant on your phone that works your apps and…
Weitere Sitemap-Seiten: VideoBildAudioKI-Musik3D-KIKurzclipsLokale LLMsKI-ChatRAG & WissenAutomatisierungProduktivitätEntwickler-ToolsDesignSchreibenSelf-HostingEinfache Installation