OpenClaw: The Fastest-Growing AI Agent Framework with 302k+ GitHub Stars
OpenClaw is the fastest-growing open-source AI agent framework on GitHub with 302k+ stars. Deploy autonomous assistants across 20+ messaging platforms including WhatsApp, Telegram, Slack, and Discord.
OpenClaw is a personal AI assistant framework that runs on your own devices and connects to the messaging platforms you already use—WhatsApp, Telegram, Slack, Discord, Signal, iMessage, and more. Created by Peter Steinberger (founder of PSPDFKit) and now backed by OpenAI, OpenClaw has become the fastest-growing open-source project in GitHub history, reaching 302,000+ stars in just 60 days. It represents a fundamental shift in how autonomous AI agents can be deployed: locally, persistently, and across multiple communication channels.
What is OpenClaw?
OpenClaw is an open-source AI agent framework that transforms Claude (or any LLM) into a persistent, autonomous assistant running on your hardware. Unlike traditional chatbots that answer questions in isolation, OpenClaw agents operate continuously, remember context between sessions, and can take autonomous actions across your digital ecosystem.
The core innovation is architectural simplicity combined with powerful integration. OpenClaw provides a local Gateway—a control plane that manages sessions, channels, tools, and events. This Gateway connects to messaging platforms (WhatsApp, Telegram, Slack, Discord, Google Chat, Signal, iMessage, BlueBubbles, IRC, Microsoft Teams, Matrix, Feishu, LINE, Mattermost, Nextcloud Talk, Nostr, Synology Chat, Tlon, Twitch, Zalo, and WebChat) and exposes a unified interface for your AI agent to interact with the world.
Created by Peter Steinberger after his $800M PSPDFKit exit, OpenClaw evolved from a weekend hack (Clawd) through iterations as Clawdbot and Moltbot before settling on its current name. The project transitioned to an independent 501(c)(3) foundation in February 2026, with OpenAI providing financial sponsorship while maintaining the MIT open-source license and community governance model.
Core Features and Architecture
Multi-Channel Inbox: OpenClaw consolidates messages from 20+ messaging platforms into a single agent workspace. Your assistant receives inbound messages from WhatsApp, Telegram, Slack, Discord, and others through a unified interface, eliminating the need to build separate integrations for each platform.
Local-First Gateway: The Gateway runs on your hardware (macOS, Linux, Windows via WSL2) as a persistent daemon. It manages authentication, session state, tool execution, and event routing. This architecture keeps your data local while enabling secure remote access via Cloudflare Tunnel or similar solutions.
Multi-Agent Routing: Route different channels, accounts, or peers to isolated agents (workspaces). A single OpenClaw instance can run multiple independent agents, each with its own configuration, memory, and tool access. This enables sophisticated multi-tenant scenarios and role-based agent deployment.
Voice Wake + Talk Mode: On macOS/iOS, OpenClaw supports wake words for hands-free activation. Android supports continuous voice mode. Integration with ElevenLabs for text-to-speech and system TTS fallback enables natural voice interactions.
Live Canvas: The macOS/iOS Canvas provides an agent-driven visual workspace. Agents can render interactive UI components using A2UI, enabling rich visual interactions beyond text-based chat.
Skills Marketplace (ClawHub): Over 800 published skills extend agent capabilities—Gmail integration, GitHub automation, Spotify control, smart home management, and more. Skills are published to ClawHub and installed via package manager, similar to npm or pip.
Model Agnosticism: OpenClaw works with Claude, GPT-4, Gemini, Grok, DeepSeek, or local models via Ollama. You choose your model provider, manage your API keys, and control your costs.
Get free AI agent insights weekly
Join our community of builders exploring the latest in AI agents, frameworks, and automation tools.
Getting Started
Prerequisites: Node ≥22, npm/pnpm/bun, and a supported LLM API key (OpenAI, Anthropic, etc.).
Installation:
npm install -g openclaw@latest
# or: pnpm add -g openclaw@latest
openclaw onboard --install-daemonThe onboarding wizard guides you through Gateway setup, workspace configuration, channel pairing, and skill installation. This is the recommended path for new users.
Quick Start:
openclaw gateway --port 18789 --verbose
# Send a message
openclaw message send --to +1234567890 --message "Hello from OpenClaw"
# Talk to the assistant
openclaw agent --message "Ship checklist" --thinking highFor development from source, clone the repository, install dependencies with pnpm, and run `pnpm gateway:watch` for auto-reload on TypeScript changes.
Real-World Use Cases
Email Triage & Automation: Your OpenClaw agent monitors Gmail, categorizes incoming messages, drafts responses, and flags urgent items. It learns your communication patterns and can autonomously handle routine emails while escalating complex messages to you.
Calendar & Meeting Management: The agent detects scheduling conflicts, autonomously reschedules meetings with appropriate notifications, and generates morning briefings with your day's agenda. It integrates with Google Calendar, Outlook, and other calendar systems.
Sales Pipeline Automation: Track leads across multiple channels, log interactions to CRM systems, send follow-up reminders, and generate pipeline reports. The agent can autonomously qualify leads based on your criteria and route them appropriately.
Smart Home Orchestration: Control lights, thermostats, and other IoT devices through natural language commands delivered via WhatsApp or Telegram. The agent can execute complex automation sequences—"prepare for bedtime" triggers lights off, thermostat adjustment, and door locks.
How It Compares
vs. LangChain: LangChain is a framework for building LLM applications with chains and agents. OpenClaw is a complete deployment platform with persistent execution, multi-channel integration, and local-first architecture. LangChain is more flexible for custom development; OpenClaw is more complete for end-user deployment.
vs. CrewAI: CrewAI specializes in role-based multi-agent collaboration for complex tasks. OpenClaw focuses on personal assistant deployment across messaging platforms. CrewAI excels at orchestrating specialized agents; OpenClaw excels at persistent, always-on personal automation.
vs. Proprietary Solutions (Google Assistant, Alexa): OpenClaw runs locally on your hardware with full code transparency. Proprietary solutions offer polish and integration but require cloud processing and data sharing. OpenClaw prioritizes privacy and control; proprietary solutions prioritize convenience.
What's Next
The OpenClaw roadmap includes enhanced security hardening (addressing the 42,000 exposed instances discovered by researchers), improved ClawHub vetting to prevent malicious skills, and expanded model support. Peter Steinberger's stated mission at OpenAI is building "an agent that even my mum can use"—suggesting focus on accessibility and safety.
The ecosystem is evolving rapidly. OpenClawd (a separate commercial entity) offers managed hosting with pre-configured security. Moltbook, a social network for AI agents, has attracted 1.5 million registered agents. Competing frameworks are launching to capture the momentum. The question isn't whether AI agents will become mainstream—OpenClaw's trajectory proves they already are. The question is whether the open-source community can maintain security and governance as adoption accelerates.
