mcp-chrome

Exposes Chrome browser capabilities to external AI clients (Claude, etc.) through a Fastify-based Node.js server (mcp-chrome-bridge) running on port 12306 that implements the Model Context Protocol. Uses bidirectional JSON-RPC over Chrome native messaging to communicate between the extension and Node.js process, with Server-Sent Events (SSE) for streaming responses and STDIO as an alternative transport mechanism for clients that don't support HTTP.

Captures user interactions (clicks, typing, navigation) in real-time and stores them as executable workflows in IndexedDB, enabling playback and modification through a visual workflow builder. Uses a transaction-based system to batch DOM mutations and event captures, with a flow data model that represents sequences of actions as nodes in a directed graph that can be executed, edited, and scheduled.

Captures and analyzes network requests made by the page, enabling workflows to wait for specific API calls, extract data from responses, or modify requests. Uses Chrome DevTools Protocol (CDP) to intercept network traffic, stores request/response metadata in the workflow context, and provides tools for conditional logic based on network events.

Delegates compute-intensive operations (transformer model inference, GIF encoding, image processing) to an offscreen document that runs in a separate execution context, preventing blocking of the main UI thread. Uses Web Workers or offscreen document APIs to parallelize computation, with message passing to communicate results back to the main extension.

Provides a command-line interface for executing recorded workflows in headless mode, enabling integration with CI/CD pipelines and server-side automation. Wraps the Node.js server with CLI commands for workflow execution, result reporting, and error handling, with support for parameterized workflows and output formatting.

Enables automation workflows to coordinate actions across multiple browser tabs and windows, with shared state management and cross-tab messaging. Uses Chrome extension message passing to synchronize state between tabs, enabling workflows that require interaction with multiple pages simultaneously or sequentially.

Enables AI agents to control the browser using visual perception by capturing screenshots, analyzing page layout, and executing actions (click, type, scroll) based on visual coordinates rather than DOM selectors. Implements a ComputerTool base class that accepts screenshot input, performs vision-based reasoning, and translates visual instructions into precise browser actions, supporting multi-step visual workflows.

Provides vector-based semantic search over page content using transformer models (ONNX Runtime) running locally in the browser's offscreen document. Embeds page text into vector space using a pre-loaded model, stores vectors in an HNSW (Hierarchical Navigable Small World) index, and enables fast approximate nearest-neighbor search for finding relevant content without keyword matching.

Implements a streaming conversation interface where AI agents (Claude, etc.) can invoke browser tools in real-time and receive results within the same conversation flow. Uses a message processing pipeline that routes tool calls to the appropriate browser automation tools, captures results, and streams them back to the agent for multi-turn reasoning without waiting for full workflow completion.

Provides an in-browser visual editor overlay (built with React/Vue) that allows editing page content without interfering with the original page functionality. Uses shadow DOM to isolate the editor UI from page styles, implements a transaction-based batch apply system to commit edits atomically, and supports undo/redo through a state management system that tracks all mutations.

Provides a visual node-based interface for constructing automation workflows where each node represents a browser action or decision point. Implements a flow data model that stores workflows as directed graphs with layout algorithms for automatic node positioning, supports conditional branching and loops, and integrates with the recording system to auto-generate nodes from captured interactions.

Injects content scripts into web pages to capture user interactions, monitor DOM changes, and execute browser automation commands at the page level. Uses a message passing architecture to communicate between content scripts and the background service worker, enabling real-time event capture (clicks, typing, navigation) and DOM mutations without blocking page execution.

Manages multiple automation projects and conversation sessions using a SQLite database with Drizzle ORM, enabling users to organize workflows, save conversation history, and switch between different automation contexts. Stores project metadata, workflow definitions, execution logs, and conversation transcripts in a structured relational schema accessible through the Node.js server.

Enables scheduling and triggering of recorded workflows based on time intervals, events, or external conditions. Implements a trigger registry that maps conditions (cron expressions, webhook events, page changes) to workflow execution, with a scheduler that manages timing and ensures reliable execution even if the browser is closed.