Arm's official MCP server for porting and tuning software on Arm — semantic docs search, codebase migration scanning, container architecture checks and LLVM-MCA assembly analysis.
ESP-IDF Tools MCP Server
Summary
Espressif's local MCP server built into idf.py — lets an AI client set the chip target, build, flash and clean an ESP-IDF project and read back config, build status and connected serial ports.
Features
- Built into idf.py — no separate server to install
- Set chip target, build, flash and clean an ESP-IDF project
- Reads project config, build status and connected serial ports
- Runs locally over stdio, keeping hardware and toolchain on your machine
- Requires ESP-IDF v6.0+ and EIM v0.8.1+
Installation
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Description
ESP-IDF Tools MCP Server is Espressif's own Model Context Protocol integration for ESP32 firmware work. It is not a separate download: the server is built into idf.py, the ESP-IDF build tool, and started with idf.py mcp-server.
Embedded development has an awkward gap for coding agents. An agent can write firmware perfectly well, but it cannot tell whether the code compiles for the right target, whether the board is even plugged in, or which serial port it appeared on. This server closes that loop by exposing the build system and the attached hardware as MCP tools and resources.
Tools
set_target(target)— select the chip targetbuild_project()— compile the projectflash_project(port)— flash firmware to a connected deviceclean_project()— remove build artifacts
Resources
project://config— the project's configurationproject://status— build status and produced artifactsproject://devices— serial ports currently connected
Requirements
ESP-IDF v6.0 or later with the mcp feature enabled, plus ESP-IDF Installation Manager (EIM) v0.8.1 or later. The server runs locally over stdio, so the board, the toolchain and the agent all stay on the developer's own machine.
Related MCP Servers
Official AntV MCP server that turns a description of your data into a rendered chart — 26 visualization types from bar and line to sankey, mind-map, fishbone and geographic maps.
Context-compression MCP server that shrinks tool outputs, logs, files and RAG chunks before they reach the model, cutting 60-95% of tokens on JSON payloads.
Persistent memory for coding agents as a single Go binary — SQLite and FTS5 behind MCP, an HTTP API, a CLI and a TUI, with no Node, Python or Docker required.