Service

Installation

2025.05.11

·Web·by Anonymous

#MCP#Chrome Extension#Proxy Server

Key Points

1MCP SuperAssistant is a Chrome extension designed to integrate the Model Context Protocol (MCP) with popular AI chat platforms, providing enhanced functionality.
2Installation can be done either directly from browser extension stores (Chrome, Firefox) or manually via developer mode for advanced users.
3Full functionality requires setting up a `config.json` for MCP server configurations and running a local proxy server using Node.js and npx.

MCP SuperAssistant is a Chrome extension designed to integrate the Model Context Protocol (MCP) with various AI chat platforms, including ChatGPT, Perplexity, Gemini, Grok, and AIStudio. Its installation is detailed through three primary methods and subsequent configuration steps.The primary installation options include: Installation from Browser Extensions Store: This is the recommended and simplest method. Users can install the extension directly from the Chrome Web Store for Chrome/Chromium-based browsers (e.g., Edge, Brave, Arc) or from the Firefox Add-ons page for Firefox, by clicking "Add to Chrome" or "Add to Firefox" respectively and confirming the installation. Manual Installation (Developer Mode): This method is suited for developers or users seeking access to the latest features. It involves obtaining the extension's source code by cloning its GitHub repository or downloading a zip file and extracting it. Subsequently, users navigate to chrome://extensions/ in their browser, enable "Developer mode" via a toggle switch, and then click "Load unpacked" to select the extracted extension directory. Upon successful loading, the MCP SuperAssistant sidebar becomes visible on supported AI platforms. Following the extension installation, two crucial configuration steps are required for full functionality: Setup of config.json: A config.json file needs to be created to define the MCP servers that the extension will interact with. This file specifies server configurations under the "mcpServers" key, where each entry defines a server by a unique name (e.g., "filesystem", "notion"). Each server configuration specifies the "command" to execute, its "args" (arguments), and optional "env" (environment variables, e.g., for API tokens like NOTION_API_TOKEN). The system supports stdio, SSE (Server-Sent Events), and Streamable HTTP types of MCP servers. Existing MCP configuration files from other clients, such as Claude (~/Library/Application Support/Claude/claude_desktop_config.json on macOS or %APPDATA%\Claude\claude_desktop_config.json on Windows) and Cursor (~/.cursor/mcp.json on macOS or %APPDATA%\Cursor\mcp.json on Windows), are compatible. An example configuration for filesystem and notion servers is provided: json{ "mcpServers" : { "filesystem" : { "command" : "npx" , "args" : [ "-y" , "@modelcontextprotocol/server-filesystem" , "/path/to/other/allowed/dir" ] } , "notion" : { "command" : "npx" , "args" : [ "-y" , "@suekou/mcp-notion-server" ] , "env" : { "NOTION_API_TOKEN" : "token" } } } } Setup of MCP SuperAssistant Proxy: The extension relies on a local proxy server to securely route MCP tool calls. This proxy is essential for the extension's full functionality. Prerequisites for running the proxy include Node.js (v16 or newer) and npx, which is bundled with Node.js. The proxy is initiated via a terminal command: npx @srbhptl39/mcp-superassistant-proxy@latest --config ./config.json. This command executes the proxy application, instructing it to use the previously configured config.json file to identify and communicate with various MCP servers.The core methodology involves the browser extension communicating with a local proxy. This proxy acts as a secure intermediary, receiving MCP tool calls from the extension and dispatching them to configured MCP servers (e.g., filesystem, notion) based on the config.json. These servers, executed as separate processes via npx commands, perform specific tasks (e.g., file system access, Notion API interactions) and return results through the proxy back to the extension, thereby extending the capabilities of AI chat platforms by providing contextual data and enabling tool use.

Service

Installation

2025.05.11

·Web·by Anonymous

#MCP#Chrome Extension#Proxy Server

Key Points

1MCP SuperAssistant is a Chrome extension designed to integrate the Model Context Protocol (MCP) with popular AI chat platforms, providing enhanced functionality.
2Installation can be done either directly from browser extension stores (Chrome, Firefox) or manually via developer mode for advanced users.
3Full functionality requires setting up a `config.json` for MCP server configurations and running a local proxy server using Node.js and npx.

MCP SuperAssistant is a Chrome extension designed to integrate the Model Context Protocol (MCP) with various AI chat platforms, including ChatGPT, Perplexity, Gemini, Grok, and AIStudio. Its installation is detailed through three primary methods and subsequent configuration steps.The primary installation options include: Installation from Browser Extensions Store: This is the recommended and simplest method. Users can install the extension directly from the Chrome Web Store for Chrome/Chromium-based browsers (e.g., Edge, Brave, Arc) or from the Firefox Add-ons page for Firefox, by clicking "Add to Chrome" or "Add to Firefox" respectively and confirming the installation. Manual Installation (Developer Mode): This method is suited for developers or users seeking access to the latest features. It involves obtaining the extension's source code by cloning its GitHub repository or downloading a zip file and extracting it. Subsequently, users navigate to chrome://extensions/ in their browser, enable "Developer mode" via a toggle switch, and then click "Load unpacked" to select the extracted extension directory. Upon successful loading, the MCP SuperAssistant sidebar becomes visible on supported AI platforms. Following the extension installation, two crucial configuration steps are required for full functionality: Setup of config.json: A config.json file needs to be created to define the MCP servers that the extension will interact with. This file specifies server configurations under the "mcpServers" key, where each entry defines a server by a unique name (e.g., "filesystem", "notion"). Each server configuration specifies the "command" to execute, its "args" (arguments), and optional "env" (environment variables, e.g., for API tokens like NOTION_API_TOKEN). The system supports stdio, SSE (Server-Sent Events), and Streamable HTTP types of MCP servers. Existing MCP configuration files from other clients, such as Claude (~/Library/Application Support/Claude/claude_desktop_config.json on macOS or %APPDATA%\Claude\claude_desktop_config.json on Windows) and Cursor (~/.cursor/mcp.json on macOS or %APPDATA%\Cursor\mcp.json on Windows), are compatible. An example configuration for filesystem and notion servers is provided: json{ "mcpServers" : { "filesystem" : { "command" : "npx" , "args" : [ "-y" , "@modelcontextprotocol/server-filesystem" , "/path/to/other/allowed/dir" ] } , "notion" : { "command" : "npx" , "args" : [ "-y" , "@suekou/mcp-notion-server" ] , "env" : { "NOTION_API_TOKEN" : "token" } } } } Setup of MCP SuperAssistant Proxy: The extension relies on a local proxy server to securely route MCP tool calls. This proxy is essential for the extension's full functionality. Prerequisites for running the proxy include Node.js (v16 or newer) and npx, which is bundled with Node.js. The proxy is initiated via a terminal command: npx @srbhptl39/mcp-superassistant-proxy@latest --config ./config.json. This command executes the proxy application, instructing it to use the previously configured config.json file to identify and communicate with various MCP servers.The core methodology involves the browser extension communicating with a local proxy. This proxy acts as a secure intermediary, receiving MCP tool calls from the extension and dispatching them to configured MCP servers (e.g., filesystem, notion) based on the config.json. These servers, executed as separate processes via npx commands, perform specific tasks (e.g., file system access, Notion API interactions) and return results through the proxy back to the extension, thereby extending the capabilities of AI chat platforms by providing contextual data and enabling tool use.

View original