Browserplex
Browserplex is an MCP server for managing multiple concurrent named browser sessions built on Playwright. It supports Chromium, Firefox, WebKit, and Camoufox (stealth Firefox) browsers. Key features include isolated session management, screenshot capture with automatic LLM-optimized resizing, structured content extraction preserving semantic markup, comprehensive browser interactions (clicking, typing, JavaScript evaluation), tab management, and console/network debugging capabilities. Solves the limitation of standard Playwright MCP servers which don't support concurrent multi-user sessions.
真实 I/O · 已连通 ✓
启动 server · tools/list · 29 个工具 · browserplex下面每个工具的字段、类型、必填项，都是我们真的把 server 启动、tools/list 拉回来的原始 JSON Schema —— 真实的工具契约。注意：下方「示例调用」是按 schema 自动生成的演示，不是真实调用的返回。
session_createCreate a new named browser session
namestringrequiredtypestringheadlessbooleanheadedbooleanelectronArgsarrayexecutablePathstringcwdstringenvobject{
"tool": "session_create",
"arguments": {
"name": "…"
}
}session_listList all active browser sessions
{
"tool": "session_list",
"arguments": {}
}session_destroyClose and cleanup a browser session
namestringrequired{
"tool": "session_destroy",
"arguments": {
"name": "…"
}
}storage_saveSave browser session cookies/storage to a named file for later reuse
sessionstringrequireddomainstringrequirednamestring{
"tool": "storage_save",
"arguments": {
"session": "…",
"domain": "…"
}
}storage_loadLoad stored session cookies/storage into a new browser session
namestringrequireddomainstringrequiredstorageNamestringtypestringheadlessbooleanheadedboolean{
"tool": "storage_load",
"arguments": {
"name": "…",
"domain": "…"
}
}storage_listList all stored browser sessions
domainstring{
"tool": "storage_list",
"arguments": {
"domain": "…"
}
}storage_deleteDelete a stored browser session
domainstringrequirednamestring{
"tool": "storage_delete",
"arguments": {
"domain": "…"
}
}storage_lockAcquire a lock for a domain (use during auth flows to prevent concurrent logins)
domainstringrequired{
"tool": "storage_lock",
"arguments": {
"domain": "…"
}
}storage_unlockRelease a lock for a domain
domainstringrequired{
"tool": "storage_unlock",
"arguments": {
"domain": "…"
}
}browser_navigateNavigate to a URL
sessionstringrequiredurlstringrequired{
"tool": "browser_navigate",
"arguments": {
"session": "…",
"url": "…"
}
}browser_navigate_backNavigate back in browser history
sessionstringrequired{
"tool": "browser_navigate_back",
"arguments": {
"session": "…"
}
}browser_snapshotGet an accessibility tree snapshot with element refs for interaction. Use refs (@e1, @e2) with click/type tools instead of CSS selectors for more reliable automation.
sessionstringrequiredinteractivebooleancompactbooleanmaxDepthnumberselectorstring{
"tool": "browser_snapshot",
"arguments": {
"session": "…"
}
}browser_take_screenshotTake a screenshot of the current page (auto-resized to fit LLM limits). Optionally save the original (un-resized) PNG to disk via savePath.
sessionstringrequiredfullPagebooleanmaxDimensionnumbersavePathstring{
"tool": "browser_take_screenshot",
"arguments": {
"session": "…"
}
}browser_clickClick an element on the page. Use refs (@e1, @e2) from browser_snapshot or CSS selectors.
sessionstringrequiredselectorstringrequiredtimeoutnumber{
"tool": "browser_click",
"arguments": {
"session": "…",
"selector": "…"
}
}browser_typeType text into an input element. Use refs (@e1, @e2) from browser_snapshot or CSS selectors.
sessionstringrequiredselectorstringrequiredtextstringrequiredsubmitbooleantimeoutnumber{
"tool": "browser_type",
"arguments": {
"session": "…",
"selector": "…",
"text": "…"
}
}browser_press_keyPress a keyboard key
sessionstringrequiredkeystringrequired{
"tool": "browser_press_key",
"arguments": {
"session": "…",
"key": "…"
}
}browser_hoverHover over an element. Use refs (@e1, @e2) from browser_snapshot or CSS selectors.
sessionstringrequiredselectorstringrequiredtimeoutnumber{
"tool": "browser_hover",
"arguments": {
"session": "…",
"selector": "…"
}
}browser_dragDrag an element to another location. Use refs (@e1, @e2) from browser_snapshot or CSS selectors.
sessionstringrequiredsourceSelectorstringrequiredtargetSelectorstringrequiredtimeoutnumber{
"tool": "browser_drag",
"arguments": {
"session": "…",
"sourceSelector": "…",
"targetSelector": "…"
}
}browser_select_optionSelect an option from a dropdown. Use refs (@e1) from browser_snapshot or CSS selectors.
sessionstringrequiredselectorstringrequiredvaluestringlabelstringindexnumbertimeoutnumber{
"tool": "browser_select_option",
"arguments": {
"session": "…",
"selector": "…"
}
}browser_file_uploadUpload file(s) to a file input element. Use refs (@e1) from browser_snapshot or CSS selectors.
sessionstringrequiredselectorstringrequiredfilesarrayrequiredtimeoutnumber{
"tool": "browser_file_upload",
"arguments": {
"session": "…",
"selector": "…",
"files": []
}
}共 29 个工具，此处展示前 20 个。
综合分
效果测评卡
进行中 · 待运行真实 I/O 已解决「输入输出是什么」；下一步接 agent 真跑，量化「效果好不好」。
The DataForSEO MCP server provides AI assistants with direct access to DataForSEO's SEO and marketing analytics APIs. Built with TypeScript, it exposes tools across four main modules: SERP for search engine results, Keywords Data for keyword research, OnPage for content analysis, and DataForSEO Labs for competitive research and market analysis. The server handles authentication with the DataForSEO API, implements field filtering for focused data retrieval, and follows a modular architecture that makes it easy to extend with additional tools. This implementation is particularly valuable for digital marketers and SEO professionals who need to analyze search rankings, research keywords, evaluate competitors, and optimize content without leaving their AI assistant conversation.
This MCP server integrates Google Lighthouse web performance auditing capabilities, built by Daniel Sogl to provide automated website performance, accessibility, SEO, and best practices analysis through the Lighthouse testing framework. The implementation enables programmatic execution of Lighthouse audits against web pages, returning detailed performance metrics, optimization recommendations, and compliance scores that can be used for continuous monitoring, automated testing pipelines, or bulk website analysis workflows.
AI SEO is a free, open-source MCP server by AutomateLab that audits web pages across 13 dimensions for AI citation eligibility. It evaluates schema validation, robots.txt compliance, llms.txt standards, entity density, and technical metadata to score how likely AI assistants such as ChatGPT, Perplexity, and Claude are to cite a given page. The server requires no API keys or registration and runs fully locally.