# System Control

**Direct File System Access**

Clawd Bot distinguishes itself by its ability to directly manipulate the file system of the host machine. Unlike cloud-based AI tools that require you to upload documents to a server, Clawd Bot operates where your data lives.

The agent can create, read, update, and delete files across any directory you specify. This allows for powerful automation workflows. You can instruct Clawd Bot to reorganize your download folder, rename image files based on their content, or aggregate data from multiple text files into a single report. It handles various file formats, including code files, markdown, PDFs, and spreadsheets.

**Shell Execution**

For advanced power users, Clawd Bot offers direct access to the system shell. This feature turns natural language requests into executable terminal commands.

When you ask the agent to install a software package, check network latency, or git clone a repository, Clawd Bot translates these intents into Bash or PowerShell commands. It executes the command, captures the standard output or error logs, and analyzes the results to determine if the task was successful or if further action is required.

**Security and Sandboxing**

Granting an AI agent access to system files and the shell requires robust security measures. Clawd Bot implements a permission-based architecture.

By default, the agent operates in a restricted mode. Users must explicitly define the root directories the agent can access. Shell commands can be set to require user confirmation before execution. This ensures that while Clawd Bot is powerful, it remains strictly under your control, preventing accidental modification of critical system files or execution of dangerous scripts.


---

# 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/2-features/system-control.md?ask=<question>
```

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.