A curated list of tools, frameworks, and platforms for building agentic operating systems — the path to autonomous AI.
- Agent Frameworks & Orchestrators
- Computer-Use & Desktop Automation
- Web Agents & Browser Automation
- Voice & Conversational AI
- Visual & Creative AI
- Developer Tools & Code Assistants
- LLM Infrastructure & Model Serving
- Security & Offensive AI
- Data, Memory & Knowledge
- Productivity & Personal Assistants
- MCP & Tool Integration
- Research & Reference
Core frameworks for building, deploying, and managing multi-agent systems — the kernel of agentic operating systems.
| Name | Description | Stars |
|---|---|---|
| block/goose | On-machine AI agent that automates development tasks from start to finish with MCP support. | |
| elizaOS/eliza | Multi-agent simulation framework with Discord, Telegram, and Twitter integration for autonomous agents. | |
| n8n-io/n8n | Fair-code workflow automation platform with native AI capabilities and 400+ integrations. | |
| simstudioai/sim | Open-source platform to build and deploy AI agent workflows. | |
| FlowiseAI/Flowise | Drag-and-drop interface for building customized LLM orchestration flows and AI agents. | |
| langflow-ai/langflow | Low-code platform for building and deploying AI-powered agents and workflows. | |
| SmythOS/sre | Cloud-native runtime for building, running, and managing intelligent agentic AI. | |
| activepieces/activepieces | Open-source AI automation framework with extensive MCP server support. | |
| FoundationAgents/OpenManus | An open-source implementation of an autonomous AI agent. | |
| letta-ai/letta | Platform for building stateful agents with advanced memory that learn over time. | |
| Significant-Gravitas/AutoGPT | An experimental open-source application showcasing the capabilities of the GPT-4 language model. | |
| open-interpreter/open-interpreter | Open-source AI agent that can execute code on your computer to perform tasks. | |
| huginn/huginn | System for creating agents that monitor and act on your behalf across the web. | |
| dot-agent/nextpy | Self-modifying framework from the future — World's first Agentic Modular System (AMS). | |
| bytedance/deer-flow | Open-source SuperAgent harness that researches, codes, and creates with sandboxes, memories, tools, and subagents. | |
| NousResearch/hermes-agent | Adaptive AI agent platform built on the Hermes model family that evolves with usage. | |
| rowboatlabs/rowboat | Open-source AI coworker with persistent memory for long-running collaborative tasks. | |
| accomplish-ai/accomplish | Open-source AI coworker that lives on your desktop and handles complex multi-step tasks autonomously. | |
| AFK-surf/open-agent | Open-source alternative to Claude Agent SDK, ChatGPT Agents, and Manus. | |
| victor36max/shire | Persistent workspaces for AI agent teams with inter-agent mailboxes, shared drive, and full context preservation. | |
| zenml-io/kitaru | Durable execution layer for AI agents with checkpoints, replay, resume, wait, and memory — no graph DSL required. |
Agents that control desktops, interact with operating systems, and automate computer tasks — the desktop environment layer.
| Name | Description | Stars |
|---|---|---|
| HeyPuter/puter | The Internet Computer — free, open-source, and self-hostable cloud desktop operating system. | |
| simular-ai/Agent-S | Open agentic framework designed to use computers just like a human. | |
| bytedance/UI-TARS-desktop | Open-source multimodal AI agent stack for desktop automation. | |
| bytebot-ai/bytebot | Self-hosted AI desktop agent that automates computer tasks via natural language. | |
| accomplish-ai/coworker | Open-source AI coworker that lives on your desktop. | |
| Crosstalk-Solutions/project-nomad | Self-contained, offline survival computer packed with critical tools, knowledge, and AI. | |
| trycua/cua | Open-source infrastructure for Computer-Use Agents with sandboxes, SDKs, and benchmarks. | |
| elder-plinius/OBLITERATUS | OBLITERATE THE CHAINS THAT BIND YOU. | |
| holaboss-ai/holaOS | Your super agent for work: local-first, learn your working context in minutes and never forget it. | |
| autonomous-ai/autonomous-computer | Build your own Personal AI Computer. | |
| skalesapp/skales | Local-first desktop AI agent; runs autonomously in the background, fully offline via Ollama or 15+ providers. |
Browser control, web scraping, and internet interaction agents — the browser layer.
| Name | Description | Stars |
|---|---|---|
| browseros-ai/BrowserOS | Open-source agentic browser serving as an alternative to proprietary AI browsing tools. | |
| browser-use/web-ui | Web interface for running and managing AI agents directly in your browser. | |
| nanobrowser/nanobrowser | Open-source Chrome extension for AI-powered web automation with multi-agent workflows. | |
| esinecan/agentic-ai-browser | AI-driven web automation agent utilizing Playwright for intelligent decision-making. | |
| unclecode/crawl4ai | Open-source, LLM-friendly web crawler and scraper for AI data gathering. | |
| MinorJerry/WebVoyager | End-to-end web agent framework powered by large multimodal models. | |
| browserless/browserless | Headless browser deployment platform optimized for Docker and cloud environments. | |
| monteslu/vibe-eyes | MCP server that enables LLMs to see and interact with browser-based applications. | |
| browser-use/browser-harness | Self-healing harness that enables LLMs to complete any browser task. | |
| Panniantong/Agent-Reach | Give your AI agent eyes to see the entire internet — Twitter, Reddit, YouTube, GitHub, and more. |
Text-to-speech, speech-to-text, voice assistants, and real-time audio — the audio subsystem.
| Name | Description | Stars |
|---|---|---|
| OHF-Voice/piper1-gpl | Fast and local neural text-to-speech engine for low-latency applications. | |
| Liquid4All/liquid-audio | Advanced speech-to-speech audio models developed by Liquid AI. | |
| RVC-Boss/GPT-SoVITS | Powerful few-shot voice cloning and text-to-speech model training framework. | |
| multimodal-art-projection/YuE | Open full-song music generation foundation model similar to Suno.ai. | |
| resemble-ai/chatterbox | State-of-the-art open-source text-to-speech engine for realistic voices. | |
| microsoft/VibeVoice | Open-source frontier voice AI for advanced audio synthesis. | |
| nazdridoy/kokoro-tts | CLI-based text-to-speech tool utilizing the Kokoro model for multiple languages. | |
| FunAudioLLM/Fun-Audio-Chat | Large audio language model built for natural, low-latency voice interactions. | |
| nari-labs/dia | TTS model capable of generating ultra-realistic dialogue in a single pass. | |
| Blaizzy/mlx-audio | Text-to-speech, speech-to-text, and speech-to-speech library built on Apple's MLX framework. | |
| KittenML/KittenTTS | State-of-the-art TTS model under 25MB — ultra-compact, high-quality voice synthesis. | |
| HumeAI/tada | Open-source speech language model for expressive, emotionally-aware audio generation. | |
| NVIDIA/Audio2Face-3D-Samples | Service to convert audio to facial blendshapes for lipsync and real-time facial performances. | |
| livekit/livekit | End-to-end realtime stack for connecting humans and AI with low-latency audio/video infrastructure. | |
| livekit/agents | Framework for building realtime voice AI agents with low-latency audio and video pipelines. | |
| isair/jarvis | 100% private AI voice assistant that lives on your computer — works offline, remembers everything. |
Image generation, video creation, 3D modeling, and visual manipulation — the graphics subsystem.
| Name | Description | Stars |
|---|---|---|
| invoke-ai/InvokeAI | Leading creative engine for Stable Diffusion models to generate professional visual media. | |
| krea-ai/realtime-video | Open-source model for high-quality, realtime AI video generation. | |
| hpcaitech/Open-Sora | Democratizing efficient video production through open-source video generation models. | |
| MamtaRajpurohit/Cutalyst | Automated video editing AI that cuts, syncs, and subtitles with one click. | |
| gyoridavid/short-video-maker | Creates short videos for TikTok, Instagram Reels, and YouTube Shorts using MCP and REST API. | |
| shrimbly/node-banana | Free and open-source node-based generative workflow platform. | |
| VAST-AI-Research/TripoSR | Fast 3D object reconstruction from a single image using state-of-the-art AI. | |
| Sanster/IOPaint | AI-powered image inpainting tool for removing or replacing objects in photos. | |
| XingangPan/DragGAN | Interactive point-based manipulation for precise control over generative images. | |
| SkyworkAI/SkyReels-V2 | Generative model designed for creating infinite-length AI films. | |
| NVlabs/Sana | Efficient high-resolution image synthesis using Linear Diffusion Transformers. | |
| Tencent-Hunyuan/HunyuanVideo-1.5 | Leading lightweight video generation model for high-quality output. | |
| duixcom/Duix-Avatar | Open-source toolkit for AI avatar creation and digital human cloning. | |
| imlixinyang/FlashWorld | High-quality 3D scene generation framework that works within seconds. | |
| PKU-YuanGroup/Helios | Real real-time long video generation model for endless streaming video synthesis. | |
| appletea233/EditThinker | Iterative reasoning framework that unlocks step-by-step thinking for any image editor. | |
| liuwei283/RealWonder | Real-time physical action-conditioned video generation model. | |
| maifoundations/Streamo | Streaming video instruction tuning framework for continuous video understanding and generation. | |
| lightningpixel/modly | Desktop app to generate 3D models from images using local AI — runs entirely on your GPU. |
Code editors, coding agents, and development tools — the IDE layer.
| Name | Description | Stars |
|---|---|---|
| voideditor/void | Open-source AI-powered code editor designed for agentic development. | |
| HKUDS/DeepCode | Open agentic coding framework for paper-to-code and web development tasks. | |
| code-yeongyu/oh-my-openagent | The best agent harness — batteries-included agent that codes like you. | |
| firecrawl/open-lovable | Chat with AI to clone and recreate any website as a modern React app in seconds. | |
| dyad-sh/dyad | Local, open-source AI app builder for power users — v0/Lovable/Replit/Bolt alternative. | |
| sipeed/picoclaw | Tiny, fast, and deployable anywhere — automate the mundane and unleash your creativity. | |
| generalaction/emdash | Open-source agentic development environment to run multiple coding agents in parallel with any provider. | |
| Mega4alik/ollm | Lightweight local LLM management and interaction toolkit. | |
| ValueCell-ai/ClawX | Desktop app providing a graphical interface for OpenClaw AI agents without using the terminal. | |
| vivekchand/clawmetry | Real-time observability dashboard for OpenClaw AI agents to visualize agent thinking. | |
| colbymchenry/codegraph | Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, and more. |
Model hosting, fine-tuning, API gateways, and inference optimization — the driver layer.
| Name | Description | Stars |
|---|---|---|
| BerriAI/litellm | Python SDK and proxy server to call 100+ LLM APIs in a unified OpenAI format. | |
| mudler/LocalAI | Self-hosted, local-first open-source alternative to OpenAI and Claude APIs. | |
| unslothai/unsloth | Ultra-fast fine-tuning and reinforcement learning framework for LLMs. | |
| AlexsJones/llmfit | Discover hundreds of models across providers with one command to find what runs on your hardware. | |
| jingyaogong/minimind | Educational project for training small-parameter GPT models from scratch. | |
| stepfun-ai/Step-3.5-Flash | Fast, sharp, and reliable agentic intelligence model optimized for speed and accuracy. | |
| headroomlabs-ai/headroom | Compress tool outputs, logs, files, and RAG chunks before they reach the LLM — 60-95% fewer tokens. |
Penetration testing, red teaming, vulnerability scanning, and security tools — the security layer.
| Name | Description | Stars |
|---|---|---|
| KeygraphHQ/shannon | Autonomous AI pentester for web applications and APIs — analyzes source code, identifies attack vectors, and executes real exploits. | |
| vxcontrol/pentagi | Fully autonomous AI agents system capable of performing complex penetration testing tasks end-to-end. | |
| samugit83/redamon | AI-powered agentic red team framework that automates offensive security operations from recon to post-exploitation. | |
| beelzebub-labs/azazel | eBPF-powered silent observer for containerized runtimes, built for malware analysis sandboxes and agentic AI monitoring. | |
| onecli/onecli | Open-source credential vault for AI agents — Rust HTTP gateway injects API keys transparently so agents never handle raw secrets. | |
| usestrix/strix | Open-source AI hackers to find and fix your app's vulnerabilities. | |
| NVIDIA/SkillSpector | Security scanner for AI agent skills — detect vulnerabilities, malicious patterns, and security risks. |
OCR, knowledge graphs, memory systems, and data infrastructure — the filesystem layer.
| Name | Description | Stars |
|---|---|---|
| CaviraOSS/PageLM | Community-driven education platform transforming study materials into interactive resources. | |
| allenai/olmocr | Toolkit for linearizing PDFs to prepare datasets for LLM training. | |
| PaddlePaddle/PaddleOCR | Comprehensive OCR toolkit supporting 100+ languages and complex layouts. | |
| datalab-to/chandra | Specialized OCR model for parsing complex tables, forms, and handwriting. | |
| docling-project/docling | Tool for converting various document formats into AI-ready structured data. | |
| bytedance/Dolphin | Document image parsing framework using heterogeneous anchor prompting. | |
| getzep/graphiti | Tool for building real-time knowledge graphs to power AI agent memory. | |
| pathwaycom/pathway | Python ETL framework for real-time analytics, stream processing, and RAG. | |
| myriade-ai/myriade | AI-native data platform for exploring and transforming data warehouses. | |
| MemTensor/MemOS | AI memory operating system for persistent skill storage in agent systems. | |
| FalkorDB/FalkorDB | Super fast graph database using GraphBLAS for GraphRAG and knowledge graphs for LLMs. | |
| MemoriLabs/Memori | SQL-native memory layer for LLMs, AI agents, and multi-agent systems. | |
| NevaMind-AI/memU | Memory system designed for 24/7 proactive agents like openclaw (moltbot, clawdbot). | |
| unbody-io/unbody | The Supabase of the AI era — modular, open-source backend for building AI-native software designed for knowledge. |
Chat interfaces, personal AI assistants, and productivity tools — the application layer.
| Name | Description | Stars |
|---|---|---|
| danny-avila/LibreChat | Enhanced ChatGPT clone with Agents, MCP, multi-model support, and enterprise-ready features. | |
| moeru-ai/airi | Self-hosted AI companion and VTuber platform capable of voice chat, game playing, and real-time interaction. | |
| openclaw/openclaw | Your own personal AI assistant. Any OS. Any Platform. The lobster way. | |
| dinoki-ai/osaurus | AI edge infrastructure for macOS. Run local or cloud models, share tools across apps via MCP. | |
| character-ai/Ovi | Experimental AI character interaction tool from the Character.ai team. | |
| qwersyk/Newelle | Newelle — Your Ultimate Virtual Assistant. | |
| huggingface/chat-ui | The open source codebase powering Hugging Face Chat with multi-model support. | |
| BasedHardware/omi | AI wearables designed for real-time transcription and speech processing. | |
| chatboxai/chatbox | Versatile AI client supporting multiple models for daily productivity. | |
| rnchg/APT | AI productivity tool featuring built-in local ChatGPT and privacy-focused models. | |
| khoj-ai/khoj | AI second brain for searching documents, the web, and building custom agents. | |
| eigent-ai/eigent | Open-source coworker desktop application for enhancing individual productivity. | |
| janhq/jan | Open-source alternative to ChatGPT that runs 100% offline on your machine. | |
| ygwyg/system | AI-powered tool for controlling your Mac remotely from any location. | |
| open-webui/open-webui | User-friendly, self-hosted web interface for interacting with various LLMs. | |
| SillyTavern/SillyTavern | Advanced LLM frontend designed for power users and roleplay. | |
| mindverse/Second-Me | Platform to train your AI self and amplify your digital presence. | |
| souls-of-waifu/souls-of-waifu | Self-hosted AI companion with voice chat and interactive capabilities. | |
| qwibitai/nanoclaw | Lightweight OpenClaw alternative that runs in containers for security — connects to WhatsApp, Telegram, Slack, Discord, Gmail. | |
| cloudflare/agentic-inbox | A self-hosted email client with an AI agent, running entirely on Cloudflare. |
Model Context Protocol servers, tool integrations, and API connectivity — the API/bus layer.
| Name | Description | Stars |
|---|---|---|
| Klavis-AI/klavis | MCP integration platform for reliable tool use by AI agents at scale. | |
| zilliztech/claude-context | Code search MCP that makes entire codebases accessible to AI agents. | |
| metorial/metorial | Connect any AI model to 600+ integrations, powered by Model Context Protocol (MCP). | |
| ttommyth/interactive-mcp | Local, cross-platform MCP server for interact with your AI agent — human in the loop. | |
| vercel-labs/json-render | Tool for dynamically rendering AI-generated JSON data into user interfaces. | |
| alibaba/OpenSandbox | General-purpose sandbox platform for AI applications with multi-language SDKs and Docker/Kubernetes runtimes. |
Research papers, curated lists, benchmarks, and reference materials — the documentation layer.
| Name | Description | Stars |
|---|---|---|
| yazinsai/srt-ai | Translate SRT files to any language, using AI magic. | |
| machinewrapped/llm-subtrans | Open-source project using LLMs to translate subtitles across formats. | |
| ngxson/smolvlm-realtime-webcam | Real-time webcam demo with SmolVLM and llama.cpp server. | |
| SkyworkAI/DeepResearchAgent | Hierarchical multi-agent system for deep research and complex task solving. | |
| virattt/dexter | Autonomous AI agent specialized in deep financial research and analysis. | |
| HKUDS/LightRAG | Simple and fast framework for retrieval-augmented generation. | |
| vikhyat/moondream | Tiny vision-language model optimized for edge devices and efficiency. | |
| facebookresearch/sam-3d-body | Inference code and models for the Segment Anything Model in 3D. | |
| NVlabs/OmniVinci | Omni-modal LLM for joint understanding of vision, audio, and language. | |
| koala73/worldmonitor | Real-time global intelligence dashboard with AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking. | |
| eugeneyan/open-llms | Curated list of open LLMs available for commercial and research use. | |
| x1xhlol/system-prompts-and-models-of-ai-tools | Collection of system prompts and models for various AI tools. |
Contributions are welcome! Please read the contribution guidelines first.