Clawd Bot is an open-source personal AI assistant that runs on the user's own machine. Users can control it through chat apps such as WhatsApp, Telegram, Discord, Slack, Signal, or iMessage, using the chat tools they are already familiar with to direct the assistant to complete tasks.
Main capabilities include:
- Multi-platform chat control: supports WhatsApp, Telegram, Discord, Slack, Signal, iMessage.
- Persistent memory: remembers the user's preferences, past conversations, and context, with no need to repeat them over and over.
- Browser control: automatically fills out forms, extracts data, and browses websites, enabling hands-free web use.
- System access: reads and writes files, executes shell commands, and interacts with the local environment.
- Skills and plugins: can be extended through community skills or built by yourself, with considerable room for customization.
- Local operation: runs 100% on the user's machine; unless the user actively requests it, data does not leave the computer.
Typical use cases include:
- Email management: automatically categorizes, summarizes, and drafts email replies.
- Calendar and scheduling: books meetings, sends invitations, and handles rescheduling.
- Flight check-in: automatically checks in 24 hours before departure and selects better seats.
- Research and summarization: browses web pages, reads articles, and provides concise summaries.
- Home automation: controls smart home devices, sets routines, and monitors the home.
- Personal finance: tracks spending, categorizes purchases, and provides insights.
Clawd Bot can connect to 50+ services and platforms. For installation, it supports macOS, Windows, and Linux; a single command can install Node.js and other required components. If Node.js is already installed, you can use npm to globally install ClawdBot; you can also clone the source code and modify it yourself.
Common questions: the tool is open-source and free to use. If using cloud models (Claude, GPT, etc.), you only need to pay the AI model API fees; local models are completely free. By default, data remains on the local machine and is only sent to external services when the user explicitly requests it (such as sending email or using a cloud AI model). Supported models include Claude, GPT-4, GPT-3.5, Gemini, Llama, Mistral, etc., and local LLMs can also be run via Ollama or LM Studio. Basic setup requires only a few clicks, and advanced features are aimed at advanced users who need more customization. ConnectingFor personal messaging apps, you only need to scan a QR code or enter a phone number; personal messaging apps do not require an API key. Permission is requested before any system operation is performed, fine-grained permissions can be configured, and you can see what will happen before execution.
In terms of limitations, the official materials do not provide specific system resource requirements, minimum supported operating system versions, a complete list of available skills and plugins, specific configuration differences for each chat platform, hardware thresholds for running local models, and other detailed information. These contents are not currently provided.