diff --git a/docs/cli-help.md b/docs/cli-help.md
new file mode 100644
index 000000000..3ba44e655
--- /dev/null
+++ b/docs/cli-help.md
@@ -0,0 +1,230 @@
+
+# Agent-Browser CLI Help
+
+## Overview
+Agent-Browser is a browser automation CLI for AI agents. This documentation provides an overview of the available commands and their usage.
+
+---
+
+## Core Commands
+
+### Navigation
+- **open/navigate/goto <url>**
+  Navigate to a URL. Supports relative URLs (converted to absolute) and special protocols (http, https, about, data, file, chrome-extension, chrome).
+
+- **back**
+  Navigate back in browser history.
+
+- **forward**
+  Navigate forward in browser history.
+
+- **reload**
+  Reload the current page.
+
+---
+
+### Interaction
+- **click <selector> [--new-tab]**
+  Click an element by selector. Optionally open in a new tab.
+
+- **dblclick <selector>**
+  Double-click an element by selector.
+
+- **fill <selector> <text>**
+  Fill a form field with text.
+
+- **type <selector> <text>**
+  Type text into an element.
+
+- **hover <selector>**
+  Hover over an element.
+
+- **focus <selector>**
+  Focus an element.
+
+- **check <selector>**
+  Check a checkbox or radio button.
+
+- **uncheck <selector>**
+  Uncheck a checkbox or radio button.
+
+- **select <selector> <value...>**
+  Select an option in a dropdown.
+
+- **drag <source> <target>**
+  Drag an element from source to target.
+
+- **upload <selector> <files...>**
+  Upload files to an input element.
+
+- **download <selector> <path>**
+  Download a file from a specified selector to a path.
+
+---
+
+### Keyboard
+- **press/key <key>**
+  Press a key.
+
+- **keydown <key>**
+  Press down a key.
+
+- **keyup <key>**
+  Release a key.
+
+- **keyboard <type|inserttext> <text>**
+  Type text using keyboard actions.
+
+---
+
+### Scroll
+- **scroll [direction] [amount] [--selector <sel>]**
+  Scroll the page. Direction defaults to "down" and amount defaults to 300 pixels.
+
+- **scrollintoview/scrollinto <selector>**
+  Scroll an element into view.
+
+---
+
+### Wait
+- **wait --url <pattern>**
+  Wait until a URL matches a pattern.
+
+- **wait --load <state>**
+  Wait for network load state (e.g., "networkidle").
+
+- **wait --fn <expression>**
+  Wait for a JavaScript expression to evaluate to true.
+
+---
+
+### Get Information
+- **get <text|html|value|attr|url|title|count|box|styles|cdp-url> [args...]**
+  Retrieve information from the page:
+  - `text`: Get text content of an element.
+  - `html`: Get HTML content of an element.
+  - `value`: Get input value.
+  - `attr`: Get attribute value.
+  - `url`: Get current URL.
+  - `title`: Get page title.
+  - `count`: Count elements matching a selector.
+  - `box`: Get bounding box of an element.
+  - `styles`: Get styles of an element.
+  - `cdp-url`: Get CDP URL.
+
+- **is <visible|enabled|checked> <selector>**
+  Check if an element is visible, enabled, or checked.
+
+---
+
+### Mouse
+- **mouse <move|down|up|wheel> [args...]**
+  Perform mouse actions:
+  - `move <x> <y>`: Move mouse to coordinates.
+  - `down`: Press mouse button.
+  - `up`: Release mouse button.
+  - `wheel <deltaY> [deltaX]`: Scroll with mouse wheel.
+
+---
+
+### Set Configuration
+- **set <viewport|device|geo|offline|headers|credentials|media> [args...]**
+  Set various configurations:
+  - `viewport <width> <height> [scale]`: Set viewport dimensions.
+  - `device <name>`: Set device configuration.
+  - `geo <latitude> <longitude>`: Set geolocation.
+  - `offline`: Set offline mode.
+  - `headers <json>`: Set HTTP headers.
+  - `credentials <username> <password>`: Set authentication credentials.
+  - `media <dark|light|reduced-motion>`: Set media preferences.
+
+---
+
+### Network
+- **network <route|unroute|requests|request|har> [args...]**
+  Manage network operations:
+  - `route <url> [--abort|--body <json>]`: Route requests.
+  - `unroute`: Unroute requests.
+  - `requests [--clear|--filter <pattern>|--type <type>|--method <method>|--status <status>]`: List requests.
+  - `request <requestId>`: Get request details.
+  - `har <start|stop> [path]`: Start/stop HAR recording.
+
+---
+
+### Storage
+- **storage <local|session> [get|set|clear] [key] [value]**
+  Manage browser storage:
+  - `get`: Retrieve a value.
+  - `set <key> <value>`: Set a value.
+  - `clear`: Clear storage.
+
+---
+
+## Common Options
+- `--session <name>`: Specify a session name.
+- `--json`: Output JSON.
+- `--headed`: Run in headed mode.
+- `--debug`: Enable debug mode.
+- `--headers <json>`: Set HTTP headers.
+- `--provider <name>`: Specify a provider.
+- `--device <name>`: Specify a device.
+
+---
+
+## Examples
+```bash
+# Navigate to a URL
+agent-browser open https://example.com
+
+# Click an element
+agent-browser click "#submit-button"
+
+# Fill a form field
+agent-browser fill "#username" "myusername"
+
+# Wait for a URL pattern
+agent-browser wait --url "**/dashboard"
+
+# Set viewport
+agent-browser set viewport 1200 800
+
+# Get page title
+agent-browser get title
+```
+
+---
+
+## Installation
+To install and use Agent-Browser, follow these steps:
+
+1. Clone the repository:
+   ```bash
+   git clone https://github.com/dextonai/agent-browser.git
+   ```
+
+2. Build the project:
+   ```bash
+   cd agent-browser
+   npm run build:all-platforms
+   ```
+
+3. Run the CLI:
+   ```bash
+   ./bin/agent-browser
+   ```
+
+---
+
+## Additional Scripts
+- `npm run version:sync`: Sync version information.
+- `npm run build:native`: Build native binaries.
+- `npm run build:linux`: Build for Linux.
+- `npm run build:macos`: Build for macOS.
+- `npm run build:windows`: Build for Windows.
+- `npm run build:all-platforms`: Build for all platforms.
+- `npm run release`: Create a release.
+- `npm run postinstall`: Post-installation tasks.
+- `npm run changeset`: Manage changesets.
+- `npm run ci:version`: CI versioning.
+- `npm run ci:publish`: CI publishing.
+- `npm run build:dashboard`: Build dashboard.