Usage Guide

Simply send a message to your connected channel (WhatsApp, Telegram, etc.) and OpenClaw will respond. The assistant understands natural language and can handle complex requests.

You: "What's on my calendar today?"
OpenClaw: [Checks your calendar and responds with today's events]

You: "Send an email to john@example.com about the meeting"
OpenClaw: [Composes and sends the email]

You: "Check the weather and remind me to bring an umbrella if it's raining"
OpenClaw: [Checks weather, sets reminder if needed]

One of OpenClaw's most powerful features is its ability to create its own skills. Simply ask it to add functionality, and it will create the necessary code and configurations.

You: "I want you to be able to transcribe audio messages using Whisper"
OpenClaw: [Creates a skill for audio transcription using Whisper API]

You: "Give yourself support for generating images with Google's Nano Banana Pro"
OpenClaw: [Adds image generation capability and configures credentials]

OpenClaw can:

• Research APIs and documentation
• Create skill files in your workspace
• Configure credentials securely
• Test and verify the new functionality

Set up automated tasks and workflows using cron jobs, webhooks, and Gmail Pub/Sub.

You: "Check my RSS feed every few hours and create a new Todoist project 
when a new issue appears"
OpenClaw: [Sets up cron job and creates the automation]

You: "Send me a daily briefing every morning at 8 AM with my calendar, 
tasks, and weather"
OpenClaw: [Creates daily briefing cron job]

Automation capabilities include:

• Cron jobs for scheduled tasks
• Webhooks for external triggers
• Gmail Pub/Sub for email triggers
• Wakeups for scheduled agent activation

OpenClaw can modify its own configuration and improve itself based on your needs.

You: "Give yourself a profile picture that combines a crustacean with Navi from Zelda"
OpenClaw: [Generates image, sets as profile picture]

You: "Create an infographic describing your structure and features"
OpenClaw: [Scans its own directory, creates infographic using image generation]

With Voice Wake and Talk Mode enabled, you can interact with OpenClaw using voice:

• Send audio messages via Telegram, WhatsApp, etc.
• Receive voice responses
• Continuous conversation mode
• Multilingual support (e.g., Italian and English)

You: [Sends audio message] "Set up voice responses with ElevenLabs"
OpenClaw: [Researches ElevenLabs API, creates test voices, asks you to choose]

You: [Chooses voice]
OpenClaw: [Configures voice for future audio replies]

Use session tools to coordinate work across multiple agents:

• sessions_list — Discover active sessions (agents) and their metadata
• sessions_history — Fetch transcript logs for a session
• sessions_send — Message another session; optional reply-back ping-pong + announce step

This allows agents to collaborate and share information without you needing to switch between chat surfaces.

• Be specific: Clear requests get better results
• Reverse prompt: Ask the agent what to do next (e.g. “Based on what you know about me and my goals, what’s the next best task we can work on?”) instead of only giving orders—you often get better, more contextual suggestions. Great when the bot is idle.
• Run tasks through OpenClaw first: For any recurring computer task, ask OpenClaw how it could be done faster or automated before you do it manually.
• Iterate: Refine requests based on responses
• Use context: Reference previous conversations
• Review skills: Check created skills in your workspace
• Monitor usage: Use /usage to track token consumption
• Security: Review DM policies and sandbox settings

More productivity and setup tips: Tips & best practices.