OpenClaw: Complete Installation Guide & Command Reference

OpenClaw is an open-source AI assistant framework that connects language models to messaging platforms. This guide covers everything from installation to daily usage.

---

Installation Methods

Method 1: Docker (Recommended)

docker pull ghcr.io/openclaw/openclaw:latest
docker run -d \
  --name openclaw \
  -v $PWD/config:/app/config \
  -v $PWD/data:/app/data \
  ghcr.io/openclaw/openclaw:latest

Method 2: NPM

npm install -g openclaw
openclaw init
openclaw start

Method 3: From Source

git clone https://github.com/openclaw/openclaw.git
cd openclaw
npm install
npm run build
npm start

---

Configuration

OpenClaw uses a YAML configuration file. Here’s a minimal setup:

# config.yaml
bot:
  name: "MyBot"
  platform: "wechat"  # or discord, telegram, slack
  
llm:
  provider: "openai"  # or anthropic, local
  model: "gpt-4"
  apiKey: "${OPENAI_API_KEY}"

plugins:
  - name: "web-search"
  - name: "code-interpreter"

Environment Variables

Variable	Description
OPENAI_API_KEY	OpenAI API key
ANTHROPIC_API_KEY	Anthropic API key
BOT_TOKEN	Platform bot token

---

Supported Platforms

Platform	Status	Notes
WeChat	✅ Supported	Via official API
Discord	✅ Supported	Bot API
Telegram	✅ Supported	Bot API
Slack	✅ Supported	Bot API
WhatsApp	🚧 Beta	Via third-party bridge

---

Plugin System

Built-in Plugins

Plugin	Function
web-search	Web search via DuckDuckGo/Google
code-interpreter	Execute Python/JavaScript code
file-manager	Read/write files
memory	Long-term conversation memory
image-gen	Image generation via DALL-E/Stable Diffusion

Installing Community Plugins

openclaw plugin install @openclaw/plugin-translator
openclaw plugin install @openclaw/plugin-calendar

---

Common Commands

Service Management

# Start in foreground
openclaw start

# Start as daemon
openclaw start -d

# Stop
openclaw stop

# Restart
openclaw restart

# View logs
openclaw logs -f

# Check status
openclaw status

Configuration

# Validate config
openclaw config validate

# Show current config
openclaw config show

# Reload config (hot reload)
openclaw config reload

Plugin Management

# List installed plugins
openclaw plugin list

# Install
openclaw plugin install <name>

# Update all
openclaw plugin update

# Remove
openclaw plugin remove <name>

---

LLM Provider Setup

OpenAI

llm:
  provider: "openai"
  model: "gpt-4"
  apiKey: "${OPENAI_API_KEY}"
  options:
    temperature: 0.7
    maxTokens: 4096

Anthropic Claude

llm:
  provider: "anthropic"
  model: "claude-3-opus-20240229"
  apiKey: "${ANTHROPIC_API_KEY}"

Local LLM (Ollama)

llm:
  provider: "ollama"
  model: "llama3:70b"
  baseUrl: "http://localhost:11434"

---

Troubleshooting

Bot not responding

1. Check openclaw logs for errors
2. Verify platform bot token is correct
3. Confirm LLM API key is valid and has credits

Rate limiting

Add rate limiting to config:

rateLimit:
  maxRequests: 60
  perMinutes: 1

Memory issues

Reduce context window or enable message pruning:

llm:
  options:
    maxContextMessages: 50

---

Summary

OpenClaw provides a unified framework for running AI assistants across multiple messaging platforms. Key strengths:

✅ Multi-platform support from a single codebase
✅ Extensive plugin ecosystem
✅ Flexible LLM backend (OpenAI, Anthropic, local models)
✅ Hot-reload configuration
✅ Docker support for easy deployment