chgjbookknohehmaocfijekhaocaanaf
Expose your web app's tools to AI assistants like Claude and Cursor via the Model Context Protocol WebMCP Bridge lets web applications expose their functionality as tools that AI assistants can discover and call. It reads tools registered on pages via the navigator.modelContext browser API and makes them available to desktop AI clients like Claude Code and Cursor through the Model Context Protocol (MCP). - Your web app registers tools using a library like webmcp-react - The extension detects those tools on activated pages - Tools from all active tabs are aggregated and exposed through a local MCP server - AI assistants connect to the server and can call your app's tools directly - Per-tab activation ("Until reload") or per-domain activation ("Always on") - Tools are namespaced by tab to avoid collisions across pages - Status indicator shows connection state and active tool count - Supports any page that implements the navigator.modelContext API 1. Install this extension 2. Add the MCP server to your AI client (see instructions: https://github.com/MCPCat/webmcp-react/blob/main/extension/README.md) 3. Open a page with registered tools and activate the extension 4. Your AI assistant can now see and call the page's tools The companion MCP server (webmcp-server) runs locally as a stdio process. Install it with: npx webmcp-server All communication stays on your machine. The extension connects to the local server over ws://127.0.0.1 and never sends data externally. Open source under the MIT license: https://github.com/mcpcat/webmcp-react
Midscene.js
Open-source SDK for automating web pages using natural language through AI. Midscene.js lets AI be your browser operator 🤖. Just describe what you want to do in natural language, and it will help you operate web pages. This Chrome extension is completely open - source. You don't need to worry about the security problem of tokens. You can also build your own Chrome extension. Github repo: https://github.com/web-infra-dev/midscene
WebMCP
WebMCP turns your browser into a programmable interface—letting AI navigate, click, and type on any website, exactly like you would. WebMCP connects AI agents to your real browser session using the Model Context Protocol (MCP). Unlike sandboxed browser automation tools, WebMCP works in YOUR browser with YOUR cookies, logins, and extensions — giving AI agents the same view of the web that you have. How it works: - Install the extension - Run the WebMCP MCP server on your machine - Connect any MCP-compatible AI agent (Claude, GPT, etc.) - The agent can navigate, click, type, read pages, and manage tabs — all in your actual browser Key features: - Accessibility tree page reading with element references - Click, type, scroll, hover by selector, text, or reference - Tab management (open, close, switch, list) - Form analysis and auto-fill - Page screenshots - Element highlighting for human oversight Built for AI agents, not human macro recorders. This is the missing bridge between AI and your real browser.
webcode bridge
webcode bridge (Browser Extension) IMPORTANT This extension is a companion for webcode gateway. You must install and start the webcode gateway extension in VS Code before using this. For most users, the recommended Edge Isolated Keepalive mode auto-loads this bridge from the VS Code extension, so manual browser-extension installation is not required. Introduction webcode bridge is the connector that links Web AI Chatbots (Gemini, ChatGPT, DeepSeek, etc.) to your local VS Code environment. It intercepts specific AI tool calls and securely forwards them to your local VS Code server, allowing the cloud AI to "see" and "operate" on your local projects. Usage Recommended Launch: Open a folder in VS Code, start webcode gateway from the status bar, and choose an AI site. The default Edge Isolated Keepalive mode loads this bridge automatically. Manual Browser Modes: If you use regular Chrome/Edge, the system default browser, or user-profile keepalive mode, install this browser extension manually first. Auto Connect: Open Gemini or other supported AI pages from webcode. The extension will automatically detect and connect to the local service (the icon will turn green). Start Chatting: Open a new chat, type your actual request first, then add /webcode or @webcode at the end of the same message. When webcode asks whether to add the initialization prompt, choose Add or press Enter. webcode replaces the trigger word with the initialization prompt, then you can review and send the message yourself. Troubleshooting: If the icon is red or gray, click the icon to view detailed troubleshooting steps. Get VS Code Extension Search in VS Code Marketplace: webcode gateway
WebMCP Inspector
Developer tool for inspecting, testing, and debugging WebMCP tools WebMCP Inspector is a developer-focused Chrome extension that makes WebMCP tooling visible and actionable inside the browser. It detects WebMCP APIs on the current page, lists available tools, shows input schemas, and lets you execute tools instantly with JSON inputs. You can also use AI chat (Gemini, OpenAI, Anthropic, or Ollama) to run multi-step tool workflows faster. Build and debug WebMCP integrations without guesswork Validate tool schemas and arguments before production use Quickly reproduce and diagnose tool execution issues Refresh and inspect dynamically registered tools in real time Speed up testing with both manual and AI-assisted execution in one UI If you’re building WebMCP-enabled apps, this extension gives you a practical inspector and test bench directly in Chrome.
