\r \r

QB X5 Use\r

\r Based on the Browser, providing comprehensive browser automation capabilities.\r \r

Installation (one-time only)\r

\r Install QQ Browser and the x5use Python package.\r \r

bash skills/qb-x5-use/scripts/install_dep.sh\r
```\r
\r
## Setup (run before each session)\r
\r
Start the X5 background service on port 18009. Must be called after Installation. If the service is already running, it exits immediately without restarting.\r
\r
```bash\r
bash skills/qb-x5-use/scripts/setup.sh\r
```\r
\r
## Commands\r
\r
### Navigation\r
```bash\r
python3 skills/qb-x5-use/scripts/go_to_url.py \x3Curl>    # Navigate to URL\r
python3 skills/qb-x5-use/scripts/go_back.py             # Go back\r
```\r
\r
### Element interaction\r
```bash\r
python3 skills/qb-x5-use/scripts/click_element.py \x3Cindex> [xpath]      # Click element by index\r
python3 skills/qb-x5-use/scripts/input_text.py \x3Cindex> \x3Ctext> [xpath]  # Fill input by index\r
python3 skills/qb-x5-use/scripts/get_dropdown_options.py \x3Cindex>       # Get dropdown options\r
python3 skills/qb-x5-use/scripts/select_dropdown_option.py \x3Cindex> \x3Ctext>  # Select dropdown option\r
```\r
\r
### Scrolling\r
```bash\r
python3 skills/qb-x5-use/scripts/scroll_down.py [amount]       # Scroll down\r
python3 skills/qb-x5-use/scripts/scroll_up.py [amount]         # Scroll up\r
python3 skills/qb-x5-use/scripts/scroll_to_text.py \x3Ctext>      # Scroll to text\r
python3 skills/qb-x5-use/scripts/scroll_to_top.py              # Scroll to top\r
python3 skills/qb-x5-use/scripts/scroll_to_bottom.py           # Scroll to bottom\r
```\r
\r
### Download\r
```bash\r
python3 skills/qb-x5-use/scripts/download_file.py \x3Cindex>      # Download file by index\r
python3 skills/qb-x5-use/scripts/download_url.py \x3Curl>         # Download file by URL\r
```\r
\r
### Content\r
```bash\r
python3 skills/qb-x5-use/scripts/get_content.py                # Get page content as Markdown\r
```\r
\r
### Wait\r
```bash\r
python3 skills/qb-x5-use/scripts/wait.py [seconds]             # Wait specified time\r
```\r
\r
## Core workflow\r
\r
1. **Navigate**: `go_to_url.py \x3Curl>`\r
2. **Read result**: Check the returned interactive elements with refs like `[0]`, `[1]`\r
3. **Interact**: Use index from the result to click, fill, select, etc.\r
4. **Re-read result**: After navigation or interaction, check new interactive elements\r
\r
## Return value\r
\r
Every command returns the current page state, including action result and interactive elements.\r
\r
### Structure\r
\r
**Action Result**\r
- Success or Failed status\r
- Target URL and Content-Type\r
\r
**Page Content**\r
\r
| Field | Description |\r
|-------|-------------|\r
| Previous page | Title and URL of the previous page |\r
| Action | Action name and parameters |\r
| Action Result | Execution result (e.g. `navigation triggered`) |\r
| Current page | Title and URL of the current page |\r
| Interactive elements | All interactive elements in the viewport, each with `[index]\x3Ctag text/>` |\r
\r
### Example output\r
\r
Navigating to Baidu:\r
\r
```bash\r
python3 skills/qb-x5-use/scripts/go_to_url.py https://www.baidu.com/\r
```\r
\r
```\r
Action result: Success! Navigated to https://www.baidu.com/, The Content-Type of the url in response headers is 'text/html; charset=utf-8'\r
\r
>>>>> Page Content\r
State of current webpage. NOTE that the following is one-time information!\r
[Start of state]\r
Previous page: 百度一下，你就知道 (https://www.baidu.com/)\r
Action: go_to_url ({"url":"https://www.baidu.com/"})\r
Action Result: navigation triggered.\r
Current page: [0] 百度一下，你就知道 (https://www.baidu.com/)\r
Interactive elements from top layer of current page inside the viewport: [Start of page]\r
[0]\x3Ca 新闻/>\r
[1]\x3Ca hao123/>\r
[2]\x3Ca 地图/>\r
[3]\x3Ca 贴吧/>\r
[5]\x3Ca 图片/>\r
[13]\x3Ctextarea />\r
[29]\x3Cbutton 百度一下/>\r
[12]\x3Ca tj_login>登录/>\r
...\r
[End of page]\r
[End of state]\r
```\r
\r
### Interactive element format\r
\r
Each element: `[index]\x3Ctag text/>`\r
\r
| Part | Description | Example |\r
|------|-------------|---------|\r
| `[index]` | Element index for `click_element`, `input_text`, etc. | `[13]` |\r
| `\x3Ctag>` | HTML element type (`a`, `button`, `textarea`, `div`, `img`, `span`) | `\x3Ctextarea>` |\r
| `text` | Display text (may be empty) | `百度一下` |\r
\r
## Example: Search on Baidu\r
\r
```bash\r
# Navigate to Baidu\r
python3 skills/qb-x5-use/scripts/go_to_url.py https://www.baidu.com/\r
# Output shows: [13]\x3Ctextarea />, [29]\x3Cbutton 百度一下/>\r
\r
# Fill search box\r
python3 skills/qb-x5-use/scripts/input_text.py 13 "搜索词"\r
\r
# Click search button\r
python3 skills/qb-x5-use/scripts/click_element.py 29\r
\r
# Check result\r
python3 skills/qb-x5-use/scripts/get_content.py\r
```\r
\r
## Example: Scroll and download\r
\r
```bash\r
# Navigate to page\r
python3 skills/qb-x5-use/scripts/go_to_url.py https://example.com/files\r
\r
# Scroll to find more content\r
python3 skills/qb-x5-use/scripts/scroll_down.py 500\r
\r
# Download file by index from interactive elements\r
python3 skills/qb-x5-use/scripts/download_file.py 5\r
\r
# Or download by direct URL\r
python3 skills/qb-x5-use/scripts/download_url.py https://example.com/file.pdf\r
```\r
\r
## Troubleshooting\r
- If an element is not found, use the returned interactive elements list to find the correct index.\r
- If the page is not fully loaded, add a `wait.py` command after navigation.\r
\r