# FAQ **1. Is it safe to give an AI access to my file system and shell?** Clawd Bot is designed with a "security first" mindset. You have full control over the agent's permissions. In the configuration file, you must explicitly define which directories the agent is allowed to access. For shell commands, you can enable a confirmation mode that requires you to approve any command before it is executed. While the tool is powerful, it never operates outside the boundaries you set. **2. Does my personal data or my prompts leave my computer?** This depends on your configuration. If you configure Clawd Bot to use a cloud-based LLM like OpenAI or Anthropic, your prompts are sent to their servers for processing. However, the agent's core logic, memory, and access to your local files and browser all run exclusively on your machine. For maximum privacy, you can configure Clawd Bot to use a local LLM via Ollama, which ensures no data ever leaves your computer. **3. How is Clawd Bot different from using ChatGPT or Claude on the web?** Web-based chatbots are language models; Clawd Bot is an agent. A chatbot can only provide you with text or code that you must then copy and execute yourself. An agent, like Clawd Bot, takes the next step: it actually executes the code, runs the shell commands, and manipulates the files on your behalf. It closes the loop between instruction and action. **4. What are the costs involved in running Clawd Bot?** The Clawd Bot software itself is free and open source. The primary cost comes from the use of a Large Language Model. If you use a commercial API like OpenAI's GPT-4, you will be billed by them based on your usage. To run the bot at zero cost, you can configure it to use a free, open-source model running on your own hardware via Ollama. **5. Do I need to own the CLAWD token to use the bot?** No. The core functionality of Clawd Bot is available to everyone without requiring any tokens. The CLAWD token is a utility token used for optional enhancements. You can stake it to unlock Pro Tier features like faster processing speeds and access to a premium skill marketplace, but it is not necessary for the standard operation of the agent. **6. The bot is not responding to my messages. What should I check first?** The most common cause is a configuration issue. First, ensure the bot process is running in your terminal without any errors. Second, double-check your .env file to confirm that your bot token is correct and that your personal user ID has been added to the TELEGRAM\_ALLOWED\_USERS list. The bot will intentionally ignore messages from any user not on this list. **7. Which Large Language Models are supported?** Clawd Bot is designed to be model-agnostic. It has built-in support for models from OpenAI, Anthropic, and any model compatible with the Ollama API. This allows you to choose the model that best fits your needs for performance, cost, and privacy. **8. Can I build my own skills or integrations?** Yes. Clawd Bot is built to be extensible. The "Skills" system allows anyone with basic programming knowledge to add new capabilities to their agent. You can write a skill to interact with a proprietary work API, control your smart home devices, or automate any other part of your digital life. Check the developer documentation for a guide on creating your first skill. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.clawdtoken.xyz/8-faq/faq.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.