Browser Automation MCP Servers
Browser control, page inspection, screenshots, web tests, and automation.
HTML/CSS to Image API
Published by GitHub account htmlcsstoimage
Hosted HCTI MCP server that lets an assistant render HTML/CSS, capture live web pages, and fill templates into images or PDFs.
AgentScrape
Published by GitHub account hshintelligence
Remote MCP server that scrapes web pages, extracts structured data and metadata, takes screenshots and runs browser workflows, paid per call in USDC.
Hwatu
Published by GitHub account hongnoul
Gives coding agents a headless WebKit browser for one-call page checks, screenshots, pixel diffs, and UI assertions.
Yad Agent Mcp
Published by GitHub account holistis
A local MCP wrapper that lets a coding agent drive Yad, a privacy-first browser companion, inside the user's own logged-in browser.
Podium Mcp
Published by GitHub account hoainho
Local stdio MCP server with 51 tools for driving iOS simulators, real iPhones and Android devices: UI automation, E2E flows, assertions, WebView and React…
Windows Computer Use Mcp
Published by GitHub account hiphopanda
Lets an assistant control a Windows desktop: move and click the mouse, type keys, take screenshots, read and write the clipboard, and open or list applications.
Computer Use
Published by GitHub account hightemp
Local MCP server that gives an assistant broad control of the desktop: mouse, keyboard, screen, windows and processes.
Hermai Cli
Published by GitHub account hermai-ai
Hermai CLI exposes the Hermai registry of website API schemas, letting an assistant look up, submit and call site endpoints from an MCP client.
Computer-Use Agents API
Published by GitHub account hcompai
Remote MCP server for running and steering H Company computer-use agents that browse the web and operate a session machine on your behalf.
MarkuprPlus
Published by GitHub account hashfunction
Desktop screen-capture MCP server that lets an AI agent grab screenshots, record screen and voice, and receive structured visual-feedback reports.
GridProof
Published by GitHub account gridproof
Renders a running frontend page with Playwright and reports spacing, grid, and tap-target violations so a coding agent can fix UI layout.
Ghost in the Droid
Published by GitHub account ghost-in-the-droid
Gives an AI assistant 62 MCP tools to see and control a real Android or iPhone screen: tap, swipe, type, launch apps, and run saved skills.
Solve
Published by GitHub account getrelove
Local MCP server that lets an AI assistant solve Cloudflare Turnstile CAPTCHAs, billed per call.
Selenium Mcp
Published by GitHub account gaforov
Selenium-based MCP server giving an AI assistant 41 tools to drive a real Chrome, Firefox or Edge browser locally, with page snapshots and multi-session…
Agent Audit
Published by GitHub account fullstackdegen
Runs Lighthouse and bounded page checks on a URL and returns prioritized fix packs a coding agent can act on.
Deepadb
Published by GitHub account fullread
Full-featured ADB MCP server — 209 tools across 49 modules, from UI to baseband.
Chrome Bridge
Published by GitHub account frsorrentino
Drives your own logged-in Chrome through a browser extension, giving an assistant 69 tools for navigation, forms, DOM inspection, debugging, audits and…
Expo Android MCP
Published by GitHub account frndchagas
A local MCP server that lets an assistant inspect and drive Android emulators or devices through ADB, including taps, swipes, text input and app launches.
Mcp Design System Extractor
Published by GitHub account freema
Local MCP server that reads a Storybook design system and lets an assistant list, search and extract component HTML, styles and theme tokens.
Quickshell MCP
Published by GitHub account franklinnolasco7
A local MCP server that grounds AI coding agents in current Quickshell, QML and Qt documentation so they can search APIs, generate and validate QML, and inspect
Flyto Core
Published by GitHub account flytohub
A local Python execution engine that gives an AI assistant hundreds of workflow modules for browser, API, file, data and crypto steps, with traced, replayable,
FlutterSDK Dusk
Published by GitHub account fluttersdk
A local stdio MCP server that lets an AI assistant drive a running Flutter app: snapshot its Semantics tree, tap, type, scroll, and take screenshots.
MCP Virtual Computer
Published by GitHub account flujo-app
Gives an assistant a persistent Docker or Fly.io computer with a terminal, file tools and an optional live Xfce desktop it can see and control.
BRYG
Published by GitHub account flexiblepackage
Lets an AI assistant drive a real Chromium browser on your machine to navigate, click, type, read pages, run OCR, and capture network traffic.