Why OpenClaw?
The open-source AI agent with the largest community
Explore the Guides
In-depth guides for every stage — from first install to production deployment
First 7 Days with OpenClaw
From zero to power user in one week — daily guide with hands-on projects
API & Webhooks
WebSocket API, webhook ingress/egress, Node.js & Python clients, integration patterns
Voice & Multimodal
Voice control, vision, image generation, real-time audio, and wearable integration
Performance Tuning
Cut costs 97% — model routing, heartbeat optimization, caching, and token reduction
Monitoring & Observability
OpenTelemetry, dashboards, alerting, health checks, and cost tracking
CI/CD & Testing
Test skills, GitHub Actions workflows, Docker deploy, Kubernetes Helm, and rollback
MCP Servers
Find, connect, and build MCP servers — one protocol for tools, data and services
Custom Channels
Build channel adapters for any messaging platform with the TypeScript SDK
Troubleshooting
Diagnostic commands, debug mode, fix recipes for every subsystem, emergency reset
ClawHub Marketplace
Browse, install, verify, publish, and secure your skill supply chain
Multi-Agent Workflows
Orchestrator-worker fleets, Workboard kanban, shared memory, and cost optimization
Security Hardening
MCP trust tiers, SOUL.md protection, incident response runbook, and audit recipes
Up and Running in Minutes
Install OpenClaw, connect a channel, and start chatting with your AI agent. The Quick Start guide gets you from zero to a working agent in under 5 minutes.
# Install OpenClaw
curl -fsSL https://openclaw.ai/install.sh | bash
# Configure your LLM provider
openclaw config set provider openrouter
openclaw config set model deepseek/deepseek-chat-v3-0324
# Connect Telegram
openclaw channel add telegram
# Launch
openclaw start