Bench
USB hardware discovery, device identification, serial communication, and diagnostics for makers and hardware engineers.
README
<p align="center"> <img src="docs/assets/bench-logo.svg" alt="Bench" width="200"> </p>
<h1 align="center">Bench</h1>
<p align="center"><strong>USB hardware discovery for your AI tools.</strong></p>
Bench is a native macOS MCP server that gives AI tools like Claude Code, Cursor, and Windsurf visibility into connected USB hardware. It identifies devices, finds serial ports, and recognizes common maker boards — so your AI assistant knows what's on your bench.
No API keys. No drivers. One command to install.
What it does
22 tools across four categories:
Discovery
| Tool | Description |
|---|---|
ping |
Health check — returns server status, version, macOS version |
list_usb_devices |
List all connected USB devices with vendor, type, speed, serial |
get_device_info |
Detailed info on a specific device by serial, location ID, or name |
identify_device |
Smart identification of 83+ known maker/dev boards |
list_serial_ports |
Enumerate serial ports with USB device matching |
hub_topology |
USB hub tree view showing port hierarchy and connections |
device_descriptors |
Full USB descriptor chain — interfaces, endpoints, class codes |
chip_detect |
Detect exact chip type of ESP32/microcontroller via esptool |
Monitoring
| Tool | Description |
|---|---|
monitor_events |
Detect USB connect/disconnect events between calls |
snapshot_state |
Capture and diff USB device state snapshots |
diagnose_device |
Query system logs for USB errors on a specific device |
power_info |
Per-device power draw, bus budgets, charging detection |
Management
| Tool | Description |
|---|---|
eject_device |
Safely unmount and eject removable storage |
tag_device |
Persistent user-defined aliases for devices |
port_reset |
Reset a USB port to recover frozen devices |
flash_firmware |
Flash firmware via esptool/dfu-util/avrdude/UF2 |
hid_send |
Send/receive raw HID reports |
Serial Communication
| Tool | Description |
|---|---|
serial_open |
Open a serial connection with configurable baud rate, data bits, parity |
serial_read |
Read available data from an open serial connection |
serial_write |
Write data or commands to an open serial connection |
serial_close |
Close an open serial connection |
serial_monitor |
Capture serial output for N seconds (boot logs, debug output) |
Features
- Device classification — automatically categorizes devices as storage, input, hub, video, serial adapter, microcontroller, or debugger
- Serial port detection — maps USB devices to their
/dev/cu.*serial ports (the #1 question makers ask) - 83+ known boards — recognizes Arduino, Raspberry Pi, ESP32, Adafruit, SparkFun, Teensy, STM32, and common USB-serial chips
- Storage info — mount points, capacity, and free space for USB drives
- USB monitoring — event tracking, state snapshots, diagnostic log queries, and power analysis
- Firmware flashing — flash ESP32, STM32, Arduino AVR, and RP2040 boards directly
- HID interaction — send and receive reports from Stream Decks, macro pads, and custom HID devices
- Serial communication — open, read, write, and monitor serial ports with configurable baud rate, data bits, parity, and stop bits
Requirements
- macOS 14+ (Sonoma or later) on Apple Silicon
- An MCP-compatible AI tool (Claude Code, Cursor, Windsurf, etc.)
- For building from source: Xcode 16.3+ / Swift 6.1+
Install
Homebrew (recommended)
brew install seayniclabs/tap/bench
From source
git clone https://github.com/seayniclabs/bench.git
cd bench
swift build -c release
codesign --force --sign - --entitlements Sources/Bench/Bench.entitlements .build/release/Bench
The binary is at .build/release/Bench.
Add to Claude Code
claude mcp add bench -- $(which bench)
Or add manually to ~/.claude.json:
{
"mcpServers": {
"bench": {
"command": "/path/to/bench",
"args": []
}
}
}
Usage
Once connected, just talk to your AI tool:
- "What USB devices are connected?"
- "What port is my Arduino on?"
- "Identify the device on /dev/cu.usbserial-2120"
- "Eject the Samsung T7"
- "Show me all storage devices"
- "Open a serial connection to /dev/cu.usbserial-2120 at 9600 baud"
- "Monitor the serial output from my ESP32 for 10 seconds"
How it works
Bench uses Apple's IOKit framework to enumerate USB devices natively on macOS. It enriches results with serial port detection (/dev/cu.* scanning), storage info (diskutil), and a built-in database of known maker boards. It communicates with AI tools over stdio using the Model Context Protocol (JSON-RPC).
AI Tool --stdio/JSON-RPC--> Bench --IOKit--> USB Device Tree
--diskutil--> Storage Info
--/dev/cu.*--> Serial Ports
--DeviceDB--> Board Recognition
No special permissions needed. IOKit USB enumeration works without entitlements from a CLI binary.
Building
swift build # debug build
swift build -c release # release build
swift test # run tests
Bench requires Swift 6.1+ and targets macOS 14+.
License
MIT
Credits
Built by Seaynic Labs.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.