Skip to main content
Glama

MCP Fetch

by matiasngf
GitHub
TypeScript
OverviewInspectNewSchemaRelated ServersReviewsScore
Need Help?View Source CodeReport Issue

MCP Fetch

A Model Context Protocol server for browser automation using Puppeteer. Control browsers, create pages, and execute arbitrary JavaScript - all through the MCP interface.

Configuration

Add this to your MCP settings configuration file:

{ "mcp-fetch": { "type": "stdio", "command": "npx", "args": [ "-y", "mcp-fetch" ] } }

Features

  • 🌐 Browser Management: Launch, list, and close browser instances
  • 📄 Page Control: Create, list, and close pages (tabs) in browsers
  • 🔧 Flexible Execution: Execute arbitrary JavaScript on pages with full Puppeteer API access
  • 📋 Comprehensive Docs: Built-in documentation via get-rules tool

Available Tools

Browser Management

launch-browser

Launch a new browser instance with customizable settings.

Parameters:

  • headless (boolean, optional): Run in headless mode. Default: false
  • width (number, optional): Browser window width. Default: 1280
  • height (number, optional): Browser window height. Default: 720

Returns:

{ "success": true, "browserId": "browser_1234567890_abc123", "message": "Browser launched successfully with ID: browser_1234567890_abc123", "config": { "headless": false, "width": 1280, "height": 720 } }
list-browsers

List all active browser instances with their details.

Parameters: None

Returns:

{ "success": true, "browserCount": 1, "browsers": [{ "id": "browser_1234567890_abc123", "createdAt": "2025-01-15T10:30:00.000Z", "tabCount": 2, "tabs": [{ "index": 0, "url": "https://example.com", "title": "Example Domain" }], "isConnected": true }] }
close-browser

Close a browser instance and clean up resources.

Parameters:

  • browserId (string, required): The ID of the browser to close

Returns:

{ "success": true, "message": "Browser browser_1234567890_abc123 closed successfully", "cleanedPagesCount": 2 }

Page Management

create-page

Create a new page (tab) in a specified browser instance.

Parameters:

  • browserId (string, required): The ID of the browser to create a page in

Returns:

{ "success": true, "pageId": "page_1234567890_xyz789", "browserId": "browser_1234567890_abc123", "message": "Page created successfully with ID: page_1234567890_xyz789" }
list-pages

List all active pages across all browser instances.

Parameters: None

Returns:

{ "success": true, "pageCount": 2, "pages": [{ "id": "page_1234567890_xyz789", "browserId": "browser_1234567890_abc123", "browserExists": true, "createdAt": "2025-01-15T10:31:00.000Z", "url": "https://example.com", "title": "Example Domain", "isClosed": false }] }
close-page

Close a specific page and remove it from active pages.

Parameters:

  • pageId (string, required): The ID of the page to close

Returns:

{ "success": true, "message": "Page page_1234567890_xyz789 closed successfully" }

Page Interaction

exec-page

Execute arbitrary JavaScript code on a page with full Puppeteer API access.

Parameters:

  • pageId (string, required): The ID of the page to execute code on
  • source (string, required): JavaScript code to execute. Has access to the page object

Returns:

{ "success": true, "result": "The return value from your code as a string" }

Example Usage:

// Navigate and interact with a page await page.goto('https://example.com'); await page.type('#search', 'hello world'); await page.click('#submit'); // Extract data const title = await page.title(); const results = await page.$$eval('.result', els => els.length); return { title, resultCount: results };

Documentation

get-rules

Get comprehensive documentation about this MCP server.

Parameters: None

Returns: Complete documentation including schemas, use cases, and best practices.

Use Cases

Web Development & Debugging

Launch browsers to test your web applications during development. View real-time changes, debug JavaScript, inspect elements, and test responsive designs.

// Launch a visible browser launch-browser({ headless: false }) // Create a page and navigate to your dev server create-page({ browserId: "browser_id" }) exec-page({ pageId: "page_id", source: ` await page.goto('http://localhost:3000'); const title = await page.title(); return title; ` })

Automated Testing

Run automated tests in headless mode to verify functionality, take screenshots, or perform regression testing.

// Launch headless browser for testing launch-browser({ headless: true }) // Run your test suite exec-page({ pageId: "page_id", source: ` await page.goto('https://myapp.com'); await page.click('#login'); await page.type('#username', 'test@example.com'); await page.type('#password', 'password'); await page.click('#submit'); // Wait for navigation await page.waitForNavigation(); // Check if login was successful const url = page.url(); return url.includes('/dashboard') ? 'Login successful' : 'Login failed'; ` })

Web Scraping

Extract data from websites that require JavaScript execution or complex interactions.

exec-page({ pageId: "page_id", source: ` await page.goto('https://news.site.com'); // Wait for content to load await page.waitForSelector('.article'); // Extract article data const articles = await page.$$eval('.article', elements => elements.map(el => ({ title: el.querySelector('.title')?.textContent, summary: el.querySelector('.summary')?.textContent, link: el.querySelector('a')?.href })) ); return articles; ` })

Best Practices

  1. Always close browsers when done to free up system resources
  2. Use headless mode for automation tasks to improve performance
  3. Use headed mode (headless: false) for debugging and development
  4. Store browser and page IDs to manage multiple instances
  5. Check existing browsers with list-browsers before launching new ones
  6. Handle errors gracefully - browsers may crash or become unresponsive
  7. For exec-page: Always return a value at the end of your code
  8. For exec-page: Use try-catch blocks for better error handling

Installation

The browser (Chromium) will be automatically downloaded on first use. If you encounter issues, you can manually install it:

npx puppeteer browsers install chrome

Limitations

  • Browser instances are stored in memory and will be lost if the MCP server restarts
  • Each browser instance consumes system resources (RAM, CPU)
  • Puppeteer requires Chromium to be downloaded (happens automatically)

License

MIT

This server cannot be installed

-
security - not tested
F
license - not found
-
quality - not tested

How are these scores calculated?

local-only server

The server can only run on the client's local machine because it depends on local resources.

Enables browser automation using Puppeteer through the MCP interface. Allows launching browsers, creating pages, and executing arbitrary JavaScript for web scraping, testing, and debugging tasks.

  1. Configuration
    1. Features
      1. Available Tools
        1. Browser Management
        2. Page Management
        3. Page Interaction
        4. Documentation
      2. Use Cases
        1. Web Development & Debugging
        2. Automated Testing
        3. Web Scraping
      3. Best Practices
        1. Installation
          1. Limitations
            1. License

              Related MCP Servers

              • Puppeteer MCP Server

                merajmehrabi
                A
                security
                A
                license
                A
                quality
                Enables browser automation with Puppeteer, supporting navigation, form interactions, and connection to active Chrome instances for comprehensive web page interaction.
                Last updated -
                8
                4,205
                192
                TypeScript
                MIT License
                • Apple
                • Linux

              • MCP Selenium

                angiejones
                -
                security
                A
                license
                -
                quality
                Enables browser automation using the Selenium WebDriver through MCP, supporting browser management, element location, and both basic and advanced user interactions.
                Last updated -
                975
                232
                JavaScript
                MIT License

              • Steel Puppeteer

                rdvo
                -
                security
                A
                license
                -
                quality
                A Model Context Protocol server that enables LLMs to automate web browsers using Puppeteer, allowing navigation, screenshots, form filling, and JavaScript execution in a real browser environment.
                Last updated -
                1
                JavaScript
                MIT License

              • Configurable Puppeteer MCP Server

                afshawnlotfi
                A
                security
                F
                license
                A
                quality
                A Model Context Protocol server that provides browser automation capabilities using Puppeteer with configurable options through environment variables, enabling LLMs to interact with web pages, take screenshots, and execute JavaScript in a browser environment.
                Last updated -
                7
                4
                JavaScript

              View all related MCP servers

              New MCP Servers

              MCP directory API

              We provide all the information about MCP servers via our MCP API.

              curl -X GET 'https://glama.ai/api/mcp/v1/servers/matiasngf/mcp-fetch'

              If you have feedback or need assistance with the MCP directory API, please join our Discord server