MCP Operator

# MCP Operator Requirements ## Overview The MCP Operator is a tool that provides browser automation capabilities to LLMs through the MCP (Model Control Protocol) interface. It enables AI models to control a web browser, interact with web pages, and analyze web content. ## Core Components ### Browser Operator - Manages browser instances through the OpenAI Computer Use API - Orchestrates browser automation via Playwright - Handles creation, navigation, and operation of browser instances ### MCP Server - Implements the MCP protocol for communication with LLMs - Exposes browser automation tools through a standardized JSON-RPC interface - Maintains state for browser sessions ## Functional Requirements ### Browser Management - **Create Browser**: Initialize a new browser instance with persistent state - **Navigate Browser**: Direct the browser to a specified URL - **Operate Browser**: Execute natural language instructions for browser interaction - **Close Browser**: Terminate a browser instance ### Job Management - **Get Job Status**: Retrieve the status and result of an operation by job ID - **List Jobs**: View recent browser operation jobs ### Web Interaction - **Browser Operation**: Ability to follow natural language instructions to interact with web content - **Project Persistence**: Maintain browser state across sessions with project identifiers ### Additional Playwright Operations - **Take Screenshot**: Capture the current browser viewport - **Get Console Logs**: Retrieve browser console output for debugging - **Get Console Errors**: Capture error messages from the browser console - **Get Network Logs**: Monitor network activity and request/response data - **Get Network Errors**: Track failed network requests - **Scroll To**: Programmatically scroll to specific coordinates on the page - **Click**: Interact with page elements through mouse clicks - **Type**: Input text into form fields and other input elements ### Browser Debugging Tools - **Run Accessibility Audit**: Evaluate page compliance with accessibility standards - **Run Performance Audit**: Measure page load times and optimization metrics - **Run SEO Audit**: Analyze page structure for search engine optimization - **Run NextJS Audit**: Specific auditing for NextJS applications - **Run Best Practices Audit**: Check adherence to web development best practices - **Run Debugger Mode**: Advanced debugging interface for troubleshooting - **Run Audit Mode**: Comprehensive page evaluation for multiple metrics ### User Notes - **Add Note**: Create and store notes related to browser operations ## Non-functional Requirements ### Performance - Handle browser operations with appropriate timeouts - Manage memory usage efficiently, especially for screenshots - Support concurrent browser instances ### Reliability - Gracefully handle errors in browser operations - Provide proper recovery mechanisms for failed operations - Implement safeguards against navigating to dangerous websites ### Logging - File-based logging with no console output (to preserve JSON-RPC communication) - Comprehensive error reporting - Multiple log levels for different operational needs ### Security - Session isolation between different browser instances - Secure handling of web content - Protection against malicious websites ## Development Standards ### Testing - Test-driven development approach - Integration tests with real APIs (no mocks) - Support for multi-step task testing ### Code Quality - Type annotations for all functions - Comprehensive documentation - Adherence to PEP 8 standards - Proper error handling - Modular and maintainable code structure ## Communication Protocol ### JSON-RPC Interface - Standard MCP protocol compliance - Clean stdin/stdout channels for communication - Structured error responses - Asynchronous job handling with status tracking