agent-browser-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AGENT_BROWSER_PATH | No | Path to agent-browser executable | agent-browser |
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 | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| browser_navigateA | Navigate to a URL in the browser |
| browser_go_backA | Navigate back in browser history |
| browser_go_forwardB | Navigate forward in browser history |
| browser_reloadB | Reload the current page |
| browser_clickA | Click on an element identified by selector or ref (e.g., '@e1' from snapshot) |
| browser_fillA | Clear and fill a text input field with the specified value |
| browser_typeA | Type text character by character (useful for triggering key events) |
| browser_hoverC | Hover over an element |
| browser_scrollA | Scroll the page in a direction |
| browser_selectA | Select one or more options from a dropdown |
| browser_checkB | Check a checkbox or radio button |
| browser_uncheckB | Uncheck a checkbox |
| browser_pressC | Press a keyboard key |
| browser_get_textA | Get text content from an element or the entire page |
| browser_get_htmlB | Get HTML content from an element or the entire page |
| browser_get_attributeB | Get an attribute value from an element |
| browser_get_urlB | Get the current page URL |
| browser_get_titleB | Get the current page title |
| browser_snapshotA | Get an accessibility tree snapshot of the page with element refs for AI interaction |
| browser_is_visibleB | Check if an element is visible |
| browser_is_enabledA | Check if an element is enabled |
| browser_is_checkedA | Check if a checkbox/radio is checked |
| browser_screenshotC | Take a screenshot of the page |
| browser_pdfB | Generate a PDF of the current page |
| browser_waitB | Wait for an element to appear or a specified time |
| browser_get_cookiesA | Get all cookies from the browser |
| browser_set_cookieB | Set a single cookie |
| browser_clear_cookiesB | Clear all cookies |
| browser_evaluateC | Execute JavaScript code in the browser context |
| browser_get_consoleC | Get console messages from the browser |
| browser_get_networkB | Get network requests made by the browser |
| browser_closeC | Close the browser |
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 32 tools
Most tools have distinct purposes, though browser_fill and browser_type both handle text input and could be confused without careful reading. Overall, actions and queries are clearly separated.
All tools share the browser_ prefix and use a consistent verb-based pattern (e.g., browser_navigate, browser_get_text, browser_set_cookie). This makes the tool set predictable and easy to navigate.
At 32 tools, this exceeds the threshold for well-scoped sets. While each tool has a purpose, the sheer number feels heavy and some utilities (e.g., separate is_visible/is_enabled/is_checked) could be consolidated.
The tool set covers core browser automation workflows—navigation, interaction, state extraction, cookies, and diagnostics. Missing elements like tab/window management and file uploads are notable but not critical for most tasks.