kogiQA MCP Web Browser
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@kogiQA MCP Web BrowserOpen localhost:3000, log in, add first item to cart, check badge shows 1"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
kogiQA MCP Web Browser
A Model Context Protocol (MCP) server that provides browser automation capabilities using kogiQA. This server enables LLMs to interact with web pages through natural language, bypassing the need for screenshots or visually-tuned models.
For what to use:
Debug style issues on your page
Automatically fix console errors.
Map and document the functionality of your web app
Automatically write end-to-end tests
Automate exploratory testing of your application
See example prompts Usage Examples
kogiQA MCP vs Playwright MCP
The kogiQA MCP Server provides a browser which incorporates the capabilities of the kogiQA browser control algorithm. This enables agents to interact with pages without a selector, saving time and tokens.
Installation
Auto install:
npx kogiqa-mcp@latestClaude Code
claude mcp add kogiqa-browser npx kogiqa-mcp@latestVS Code
Click one of the buttons below to install directly in vs code:
Alternatively, install via the VS Code CLI:
code --add-mcp '{"name":"kogiqa-browser","command":"npx","args":["kogiqa-mcp@latest"]}'Cursor
Alternatively, go to Cursor Settings → MCP → Add new MCP Server and enter the command
npx kogiqa-mcp@latest.
Standard configuration
For any other MCP client, add the following to your MCP configuration:
{
"mcpServers": {
"kogiqa-browser": {
"command": "npx",
"args": [
"kogiqa-mcp@latest"
]
}
}
}Usage Examples
Once the server is configured, simply talk to your agent in natural language. Because kogiQA is selector-free, you describe what you want done — not how to find the elements.
Smoke-test a user flow
Open http://localhost:3000, log in with the demo credentials, add the first product to the cart, and verify that the cart badge shows "1".Debug a style issue
Go to https://myapp.example.com/pricing and check why the "Pro" plan card overflows its container on a 1280px-wide viewport. Suggest a CSS fix.Fix console errors
Navigate to http://localhost:5173, open every page linked from the main menu, collect all console errors, and fix them in the source code.Map and document your app
Explore https://staging.example.com, follow all internal links up to two levels deep, and produce a Markdown document describing each page and its main features.Generate end-to-end tests
Walk through the sign-up flow at http://localhost:3000/signup and write Cypress end-to-end tests covering the happy path and validation errors.Exploratory testing
Do 10 minutes of exploratory testing on http://localhost:8080: try edge-case inputs in every form you find and report anything that looks broken or inconsistent.Requirements
Node.js 20 or newer
VS Code, Cursor, Windsurf, Claude Desktop, Goose, Junie or a other MCP client
This server cannot be deployed
Maintenance
Related MCP Connectors
Hosted browser for AI agents: screenshots, post-JS DOM, console, WCAG. No install, no API key.
Browser-based QA for AI-built software. Test pages with real browsers via agents.
AI-powered web automation. Navigate websites using AI agents for one page or a thousand
AI-powered web automation. Navigate websites using AI agents for one page or a thousand
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to automate and debug real Chromium browsers with capabilities like screenshots, video recording, performance analysis, visual regression testing, and OCR text extraction.13-
- AlicenseNot gradedqualityDmaintenanceEnables AI coding agents to visually interact with frontend apps by taking screenshots, clicking elements, reading console logs, and performing visual diffs.20 npm3MIT
- FlicenseNot gradedqualityCmaintenanceEnables LLMs to autonomously navigate, type, click, and extract data from web pages using Playwright, without hardcoded selectors.6,282 npm-
- AlicenseNot gradedqualityBmaintenanceEnables AI coding agents to autonomously interact with and test web applications in a real browser, providing DOM/Accessibility tree extraction, runtime telemetry, screenshot capture, and Markdown test reports.32 npm1MIT