================== /// MCP /// /// INT /// ================== [server:online] [protocol:ready]
interactive-mcp
by ttommyth
Node/TypeScript implementation of a local, cross-platform Model Context Protocol (MCP) server that lets LLM agents interact with end-users (prompts, notifications, terminal chat, etc.).
273
21
Open Source01
request_user_input
Asks the user a question and returns their answer, optionally showing predefined options
02
message_complete_notification
Sends a simple operating-system notification
03
start_intensive_chat
Initiates a persistent command-line chat session
04
ask_intensive_chat
Asks a question within an active intensive chat session
05
stop_intensive_chat
Closes an active intensive chat session
Installation
1. Prerequisites
• Node.js ≥ 18 and npm (or yarn/pnpm)
• Git
2. Clone the repository
git clone https://github.com/ttommyth/interactive-mcp.git
cd interactive-mcp
3. Install dependencies
npm install # or: yarn install / pnpm install
4. Build the TypeScript sources
npm run build
5. Start the local MCP server
npm start # default entry-point
6. (Optional) Global CLI install (if published to npm)
npm install -g interactive-mcp
interactive-mcp --help
7. Configuration
• Create/adjust a .env file in the project root for any API keys or port settings required by your AI agent workflow (see README for variables).
• By default the server listens on http://localhost:3000.
8. Testing / Dev mode
npm run dev # hot-reload development server
Documentation
License: MIT License
Updated 7/30/2025