Build an ESP32 AI voice robot with XiaoZhi and MCP: media published by 78/xiaozhi-esp32Credit: 78/xiaozhi-esp32

Build an ESP32 AI voice robot with XiaoZhi and MCP

Desk, Medium

XiaoZhi is an open-source ESP32 voice chatbot that talks to large models and controls devices through MCP; the community uses it as the brain for robot dogs and desk bots.

View build on GitHub ↗Share on X

It supports 170+ board variants, needs ESP-IDF 6.x, and lists builds like the ESP-HI low-cost robot dog. MCP tools let the model move servos or toggle hardware.

Hardware & cost: An ESP32-S3 board with mic, speaker and screen, often $15-40.

Difficulty: Medium.

Agent install prompt (paste into Claude Code, Cursor or Codex): "Clone https://github.com/78/xiaozhi-esp32, install ESP-IDF 6.x, build for my board, flash it, and add an MCP tool that moves a servo when I ask."

Source: 78/xiaozhi-esp32 on GitHub — https://github.com/78/xiaozhi-esp32 (facts checked against the project README, Oct 2026; prices vary by region).

Quick answers

What do you need to build an ESP32 AI voice robot with XiaoZhi and MCP?: An ESP32-S3 board with mic, speaker and screen, often $15-40.
How hard is it to build an ESP32 AI voice robot with XiaoZhi and MCP?: Medium.

Interfaces