# AI use cases for OpenClaw (openclaw/openclaw) Repo: [openclaw/openclaw](https://github.com/openclaw/openclaw) | Use cases: 3 | Page: https://dedupedai.com/use-cases/openclaw-openclaw/ OpenClaw is an open-source personal AI assistant that runs on your own computer and answers you in the chat apps you already use, such as WhatsApp, Telegram, Slack, Discord, Signal and iMessage, plus native apps for desktop and phone. One Gateway process holds sessions, tools and channel connections; models and agent harnesses (Claude, Codex, local models) are swappable plugins. State, memory and credentials stay on your hardware, and there is no paid tier or hosted service. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Run your own AI assistant on your laptop in 10 minutes: https://dedupedai.com/use-cases/openclaw-openclaw/run-your-own-ai-assistant-on-your-laptop/ - Message your assistant from WhatsApp, Telegram or Slack: https://dedupedai.com/use-cases/openclaw-openclaw/message-your-assistant-from-whatsapp-or-telegram/ - Swap Claude, Codex or a local model behind the same assistant: https://dedupedai.com/use-cases/openclaw-openclaw/swap-models-and-harnesses-without-rebuilding/ ## How to run your own AI assistant on your laptop in 10 minutes with OpenClaw Repo: [openclaw/openclaw](https://github.com/openclaw/openclaw) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README Install OpenClaw, let onboarding check your model and set up the Gateway, then chat with it in the Control UI. **Who it is for:** Anyone who wants a personal assistant that keeps its memory and credentials on their own machine instead of a vendor's cloud. **The problem:** Hosted assistants keep your history on someone else's servers and only live in one app. **The outcome:** A running Gateway with a working assistant you can talk to in the browser dashboard, ready to connect to chat apps. ### Steps 1. On macOS, Linux or WSL2 run `curl -fsSL https://openclaw.ai/install.sh | bash` (Windows PowerShell: `iwr -useb https://openclaw.ai/install.ps1 | iex`) 2. If you already manage Node.js, use `npm install -g openclaw@latest --allow-scripts=openclaw` instead 3. Run `openclaw onboard --install-daemon`; onboarding verifies model access, creates the workspace and configures the Gateway 4. Check it with `openclaw gateway status`, then open the Control UI with `openclaw dashboard` 5. Send a message in the Control UI to confirm the assistant answers ### Prompt to paste into your agent ```text Install openclaw/openclaw on this machine with the official installer, run openclaw onboard --install-daemon using , confirm openclaw gateway status is healthy, and tell me what the onboarding wrote to disk and where. ``` ### Caveat Risk, remote script: The one-line installer runs a script fetched from the internet; read it first or use the npm install instead. Risk, credentials: Model keys and chat app tokens are stored on your machine, so protect the account it runs under. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openclaw-openclaw/run-your-own-ai-assistant-on-your-laptop/ ## How to message your assistant from WhatsApp, Telegram or Slack with OpenClaw Repo: [openclaw/openclaw](https://github.com/openclaw/openclaw) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Connect channels to the Gateway so the same assistant answers you wherever you already chat, with pairing for unknown senders. **Who it is for:** Someone who wants to ask their assistant things from their phone without opening a laptop or a new app. **The problem:** A desktop-only assistant is useless when you are away from the computer. **The outcome:** The assistant replying in your chat apps, with strangers held back until you approve a pairing request. ### Steps 1. Finish onboarding so the Gateway is running 2. Follow the getting started guide's channel setup for WhatsApp, Telegram, Slack, Discord, Signal or iMessage 3. Send the assistant a direct message from the connected app 4. Unknown senders are paired by default; approve a request with the `openclaw pairing approve` command only for people you trust 5. Read the security guide before connecting other users or exposing the Gateway remotely ### Prompt to paste into your agent ```text Help me connect my running openclaw/openclaw Gateway to . Walk me through the channel setup from the docs, keep DM pairing on, and list every token or credential I am adding and where it is stored. ``` ### Caveat Treat inbound messages as untrusted input: tools run on the host for the main session unless you configure sandboxing. Risk, credentials: Each channel adds a bot token or linked session on your machine; anyone with that machine can act as your assistant. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openclaw-openclaw/message-your-assistant-from-whatsapp-or-telegram/ ## How to swap Claude, Codex or a local model behind the same assistant with OpenClaw Repo: [openclaw/openclaw](https://github.com/openclaw/openclaw) | Difficulty: Intermediate | Time: 20 minutes | Basis: From the README Models and agent harnesses are plugins, so you can change what powers the assistant without touching your channels or memory. **Who it is for:** A tinkerer who wants to compare providers or move to a local model to save cost or keep prompts private. **The problem:** Changing AI provider usually means a new app, new history and new setup. **The outcome:** The same assistant, memory and channels running on a different model provider or harness. ### Steps 1. Open the model providers docs and pick a hosted or local provider 2. Add the provider and make it the default for the assistant 3. Add tools, skills and plugins from the docs or ClawHub as needed 4. Set `update.checkOnStart: false` if you want no version check or feature statistics sent ### Prompt to paste into your agent ```text On my openclaw/openclaw install, switch the assistant's model to , keep my existing channels and memory, and confirm with a test message which model answered. ``` ### Caveat Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openclaw-openclaw/swap-models-and-harnesses-without-rebuilding/