wrave-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| wrave_list_tabsA | List all open tabs in Brave, including tab ID, URL, title, and active state. |
| wrave_open_tabA | Open a new tab with the specified URL. |
| wrave_close_tabA | Close an open tab by its ID. |
| wrave_focus_tabB | Bring a tab to the foreground and activate it. |
| wrave_reload_tabC | Reload the specified tab. |
| wrave_get_tab_stateB | Read the active tab state including title, URL, viewport dimensions, and scroll offset. |
| wrave_get_dom_treeB | Extract clean, LLM-friendly DOM tree with bounding boxes [x, y, w, h] and interactive element metadata. |
| wrave_get_accessibility_treeA | Extract the full accessibility tree (AXTree) for semantic screen-reader level inspection. |
| wrave_take_screenshotA | Capture a screenshot of the tab viewport or full page as base64 PNG/JPEG. |
| wrave_clickB | Dispatch mouse click to element selector or pixel coordinates. |
| wrave_double_clickC | Dispatch double-click event to element selector or coordinates. |
| wrave_type_textC | Type text into an input or textarea. |
| wrave_press_keyB | Send a keyboard key press (e.g. Enter, Escape, Tab, Backspace). |
| wrave_scroll_pageC | Scroll the active tab viewport by deltaX and deltaY. |
| wrave_execute_scriptB | Execute arbitrary JavaScript expression in the tab context. |
| wrave_get_cookiesC | Retrieve cookies for the tab or domain. |
| wrave_print_to_pdfB | Export the current page as a PDF document. |
| wrave_get_console_logsA | Retrieve intercepted console logs for the tab. |
| wrave_get_snapshotA | Instant page snapshot with indexed interactive elements (@1, @2, etc.), title, and URL. |
| wrave_click_and_readB | Click an element (by selector, @ref like @1, or x/y) and immediately return the updated page snapshot in 1 turn. |
| wrave_type_and_submitB | Type text into an input or contenteditable field and submit with Enter key in 1 turn. |
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 21 tools
Most tools have clearly distinct purposes (tab management, content extraction, interaction, utilities). Some overlap exists between get_dom_tree, get_accessibility_tree, and get_snapshot, and compound tools like click_and_read combine actions but descriptions clarify their use.
All tools follow the consistent pattern 'wrave_' + verb + noun (e.g., list_tabs, close_tab, take_screenshot, execute_script). The naming is uniform and predictable, with no style mixing.
At 21 tools, the server is on the heavy side, but the scope (browser automation) justifies the breadth. It covers tab management, page inspection, interaction, and auxiliary functions without being excessive.
Core browser automation workflows are well covered: tabs (open, close, focus, reload), content extraction (DOM, AX, screenshot, snapshot), interaction (click, type, keys, scroll), and utilities (cookies, PDF, console). Minor gaps like hover, back/forward navigation, or wait-for-element are absent but not critical.