MCP Client Chatbot

What is MCP Client Chatbot?

MCP Client Chatbot is an open-source chatbot template built with Next.js and the Vercel AI SDK that integrates external tools via the Model Context Protocol (MCP). It runs as a web application on localhost, Docker, or Vercel, and is designed for developers and users who want a seamless, tool‑augmented chat experience.

How to use MCP Client Chatbot?

Install dependencies with pnpm i, copy the auto‑generated .env file, and fill in at least one LLM provider API key (e.g., OPENAI_API_KEY). For local use, start PostgreSQL with pnpm docker:pg, run migrations (pnpm db:migrate), then start the dev server (pnpm dev). Alternatively, use Docker Compose (pnpm docker-compose:up). Once running, add MCP servers via the built‑in configuration interface and begin chatting.

Key features of MCP Client Chatbot

• MCP tool integration for browser automation, APIs, and more.
• Realtime voice assistant with MCP tool support.
• Quick tool mentions via @ and switchable presets.
• Tool Choice Mode: Auto, Manual, or None per chat.
• Temporary chat windows for quick side conversations.
• Agentic chatbot with customizable project instructions.

Use cases of MCP Client Chatbot

• Automate multi‑step browser tasks (e.g., navigate, read, close) using Playwright MCP.
• Build a voice‑activated assistant that can call tools in real time.
• Organize tools by workflow with quick‑switch presets and mentions.
• Create a custom assistant for specific projects by combining MCP servers with instructions.
• Test and iterate on MCP tools in a dedicated chat environment.

FAQ from MCP Client Chatbot

—