Chrome Debug MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| REMOTE_HOST | No | The URL of the Chrome debug port | http://localhost:9222 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| launch_browserB | 启动浏览器连接,连接到Chrome调试端口以保持登录状态 |
| close_browserC | 关闭浏览器连接 |
| navigate_toC | 导航到指定URL,智能管理标签页(相同域名复用标签页) |
| clickC | 在指定坐标位置点击 |
| type_textC | 输入文本内容 |
| scroll_downB | 向下滚动页面一个视口高度 |
| scroll_upB | 向上滚动页面一个视口高度 |
| hoverC | 将鼠标悬停在指定坐标位置 |
| resize_browserB | 调整浏览器窗口大小 |
| get_page_contentB | 获取当前页面的HTML内容 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Each tool has a clearly distinct purpose with no overlap: click, hover, type_text handle different user interactions; navigate_to, scroll_up/down manage navigation; launch_browser, close_browser, resize_browser control browser state; get_page_content retrieves content. The descriptions clearly differentiate actions like click vs hover or scroll_up vs scroll_down.
Tool names follow a consistent snake_case pattern throughout (e.g., get_page_content, navigate_to). The verb-noun structure is mostly predictable, though some tools use simple verbs like click or hover without objects, which is a minor deviation from the full verb_noun pattern seen in others like resize_browser.
With 10 tools, the count is well-scoped for a Chrome debugging server. Each tool earns its place by covering essential browser automation tasks: launching/closing, navigation, interaction (click, hover, type), scrolling, resizing, and content retrieval. This provides a balanced set without being overwhelming or too sparse.
The toolset covers core browser automation workflows well, including setup (launch_browser), navigation (navigate_to, scroll), interaction (click, hover, type_text), and teardown (close_browser). Minor gaps exist, such as no explicit tool for handling pop-ups, taking screenshots, or executing JavaScript, but agents can work around these with the provided tools for most debugging tasks.