browser-search-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BROWSER_SEARCH_HEADLESS | No | Whether to run browser in headless mode. Set to 'true' to enable headless. | false |
| BROWSER_SEARCH_LOG_LEVEL | No | Logging level. Options: DEBUG, INFO, WARNING, ERROR. | INFO |
| BROWSER_SEARCH_DEFAULT_ENGINE | No | Default search engine to use. Options: google, bing, baidu, duckduckgo. |
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
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| web_searchB | Search the web using a real browser and return structured results. Uses Chrome/Edge via CDP (same approach as browser-takeover-bridge) to navigate to the search engine, extract results from the DOM, and return them as structured JSON. When the browser-takeover extension bridge is available, searches use the extension's authenticated browser session instead of a headless browser -- giving access to logged-in content. |
| web_search_multiA | Search multiple search engines and combine results. Searches multiple engines in sequence and returns combined results grouped by engine. |
| web_search_read_pageA | Navigate to a URL and return the visible page text. Useful for reading the full content of a search result. |
| web_search_statusA | Check the current browser, bridge, and provider status. Detects available CDP browser instances, checks if the browser-takeover extension bridge is running, and reports the active provider type. Returns: JSON status information |
| web_search_discover_browsersA | Scan for browsers with CDP enabled on common ports. Checks ports 9222, 9223, 9333 for Chrome/Edge instances with remote debugging enabled. Returns: JSON list of discovered browser instances |
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 5 tools
web_search and web_search_multi are related but clearly separated by the multi-engine behavior. web_search_status and web_search_discover_browsers overlap somewhat in browser inspection, but their purposes remain distinct. web_search_read_page is unambiguous.
Most tools use the web_search_ prefix with descriptive suffixes like multi, read_page, status, and discover_browsers. The base web_search tool lacks a suffix, but the pattern remains predictable and readable overall.
Five tools is well-scoped for a browser-based search server. Search, multi-engine search, page reading, status, and browser discovery each earn their place without unnecessary overlap or bloat.
The core workflow of searching, combining results, and reading pages is well covered. Browser status and discovery support the environment side. Minor gaps like explicit browser launch/close management exist, but they are outside the apparent search-focused purpose.