Browser[X]MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_PORT | No | Port for the MCP server | 3001 |
| BROWSER_HEADLESS | No | Whether to run browser in headless mode | false |
| OPENROUTER_MODEL | No | The OpenRouter model to use | deepseek/deepseek-r1:free |
| OPENROUTER_API_KEY | Yes | Your OpenRouter API key (required for AI testing) |
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
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| extract_virtual_canvasA | Extract virtual canvas data from current page instead of taking screenshot |
| navigate_browserB | Navigate browser to a specific URL |
| input_textA | Input text into a form field using virtual canvas data |
| scroll_pageB | Scroll the page in specified direction |
| start_browserC | Start browser instance for testing |
| compare_with_screenshotA | Compare virtual canvas data size with screenshot for performance testing |
| list_navigation_elementsB | Step 1: List all available interactive elements on the page with descriptions |
| get_element_detailsC | Step 2: Get detailed information about specific element for precise action execution |
| execute_atomic_actionB | Execute an atomic action generated by get_element_details |
| get_performance_metricsA | Get Browser[X]MCP server performance metrics and statistics |
| helpA | Get detailed usage instructions and workflow examples for Browser[X]MCP |
| batch_actionsA | Execute multiple actions in batch (up to 5 actions for performance) |
| evaluate_in_pageC | Execute JavaScript code in the browser page context |
| click_element_by_xpathC | Click an element using XPath selector |
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 14 tools
Most tools target distinct actions (navigate, input, scroll, start, etc.), and the two-step element workflow is clear. However, click_element_by_xpath and execute_atomic_action both can perform clicks, and evaluate_in_page could also replicate any action, creating minor overlap. Descriptions mostly clarify but ambiguity exists in action execution tools.
Almost all tools follow snake_case verb_noun pattern, with verbs like extract, navigate, input, scroll, start, list, get, execute, evaluate, click. Inconsistencies: 'compare_with_screenshot' uses preposition structure, and 'batch_actions' uses a noun as verb, but overall readable and predictable.
14 tools is within the ideal range for a browser automation server, covering navigation, interaction, element inspection, and performance utilities without feeling bloated.
The set covers navigation, interaction, element inspection, and even performance testing, but lacks a browser shutdown/close tool and explicit waiting/synchronization operations, which are common in browser automation. Core workflows are present but the lifecycle is incomplete.