Engwe N1 Air MCP Server
Enables Claude AI to control an Engwe N1 Air e-bike via Bluetooth and IoT API, including speed limit adjustment, lock/unlock, find bike, and real-time stats.
README
š² Engwe N1 Air ā BLE Reverse Engineering + Claude MCP Server
Reverse engineered the Engwe N1 Air e-bike's Bluetooth protocol and built a Claude AI integration to control it from your Mac.
What This Does
- š Unlock speed limit via Bluetooth (up to 40 km/h)
- š Lock/unlock the bike remotely via IoT API
- š Find your bike (sound + light alarm)
- š Battery, mileage, cadence real-time stats
- š Geo-fence management
- šŖ PAS level control via Bluetooth
- š¤ Claude AI integration ā control your bike by talking to Claude
- š„ļø Mac menubar app ā one-click access to all features
How It Was Built
Step 1 ā API Discovery (Charles Proxy)
Captured network traffic from the Engwe iOS app using Charles Proxy. Found the main API endpoint at app-api.engweapp.com with JWT authentication.
Step 2 ā APK Decompile (JADX)
Decompiled the Android APK using JADX to find all internal API endpoints and BLE command codes.
jadx -d engwe-source ENGWE_2.7.2_APKPure.xapk
grep -r "iot/device" engwe-source --include="*.java" | grep -o '"[^"]*iot/device[^"]*"' | sort -u
Step 3 ā BLE Protocol Reverse Engineering
Found the complete BLE protocol in ProtocolUtil.java and BLEBikeCommand.java.
V2 Protocol Format:
456E + CMD(4) + LEN(4) + DATA + CRC16(4) + 7765
BLE UUIDs (from BluetoothRegulate.java):
Service: 00001820-0000-1000-8000-00805f9b34fb
Write: 00001822-0000-1000-8000-00805f9b34fb
Notify: 00001821-0000-1000-8000-00805f9b34fb
BLE Command Reference
All commands use V2 protocol format: 456E + CMD + LEN + DATA + CRC16 + 7765
| Command | Hex | Description |
|---|---|---|
| Speed Limit | 2A65 |
0x00=0, 0x10=10, 0x20=20, 0x30=30, 0x40=40 km/h |
| PAS Level | 010A |
0x00ā0x05 |
| Motor Current Limit | 2A66 |
Under investigation |
| Cruise Control Set | 0128 |
Under investigation |
| Wheel Diameter | 08C3 |
Under investigation |
| Factory Reset | 8049 |
ā ļø Irreversible |
| Lock State | 0108 |
Query lock |
| Power On/Off | 0107 |
01=on, 02=off |
| Auto Shutdown Time | 1000 |
Minutes |
Speed Limit Example (40 km/h):
# Index 4 * 16 = 0x40
command = "456E2A650001406A237765"
CRC16 Calculation (ModBus):
def crc16(hex_str):
data = bytes.fromhex(hex_str)
crc = 0xFFFF
for byte in data:
crc ^= byte
for _ in range(8):
crc = (crc >> 1) ^ 0xA001 if crc & 1 else crc >> 1
return format(crc, "04X")
# CRC input = CMD + LEN + DATA
# Example: crc16("2A65" + "0001" + "40") = "6A23"
IoT API Endpoints
Base URL: https://app-api.engweapp.com
| Endpoint | Description |
|---|---|
/api/app/iot/device/lockSwitch |
Lock/unlock bike |
/api/app/iot/device/findVehicle |
Sound + light alarm |
/api/app/iot/device/vibrationAlarmSwitch |
Vibration alarm + signal |
/api/app/iot/device/fenceSwitch |
Geo-fence on/off |
/api/app/iot/device/getFenceList |
List geo-fences |
/api/app/iot/device/bleSenselessSwitch |
BLE auto-unlock |
/api/app/iot/device/headlightSwitch |
Headlight + realtime PAS/charge |
/api/app/iot/device/attr/newest |
Battery SOC/SOH, mileage, cadence |
/api/app/iot/device/version |
IoT/BT/4G firmware versions |
/api/app/iot/device/attr/newest |
Real-time attributes |
/api/app/paypal/syDetail |
IoT subscription info |
/api/app/ride-data/index |
Weekly ride statistics |
Installation
Requirements
- Node.js 18+
- Python 3.9+
- Mac (for menubar app)
git clone https://github.com/SafaAslan/engwe-n1air-mcp
cd engwe-n1air-mcp
npm install
pip3 install bleak
Get Your Token
Use Charles Proxy to capture traffic from the Engwe iOS/Android app. Find the Authorization header in any request to app-api.engweapp.com.
Configure Claude Desktop
~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"engwe-n1air": {
"command": "node",
"args": ["/path/to/engwe-n1air-mcp/index.js"],
"env": {
"ENGWE_IMEI": "YOUR_IMEI",
"ENGWE_BLE_MAC": "YOUR_BLE_MAC",
"ENGWE_TOKEN": "YOUR_JWT_TOKEN"
}
}
}
}
Find Your IMEI & BLE MAC
# IMEI ā visible in Engwe app under bike settings
# BLE MAC ā run while bike is on:
python3 -c "
import asyncio
from bleak import BleakScanner
async def scan():
devices = await BleakScanner.discover(timeout=10)
for d in devices:
if d.name and 'ENGWE' in d.name.upper():
print(d.address, d.name)
asyncio.run(scan())
"
Usage
Claude AI (via MCP)
Once configured, just talk to Claude:
- "Lock my bike"
- "What's my battery level?"
- "Set speed limit to 35 km/h"
- "Show my ride stats for this week"
Menubar App (Mac)
npm install electron
./node_modules/.bin/electron tray.js
Auto-start on Boot
chmod +x setup_autostart.sh
bash setup_autostart.sh
Speed Limit Script (standalone)
python3 ble_speed3.py
Project Structure
engwe-n1air-mcp/
āāā index.js # Claude MCP server (18 tools)
āāā tray.js # Mac menubar app
āāā ble_speed3.py # Standalone speed limit script
āāā ble_scan2.py # BLE device scanner
āāā setup_autostart.sh # Mac autostart setup
āāā README.md
ā ļø Disclaimer
Unlocking the speed limit above 25 km/h is illegal on public roads in the EU. This project is for educational and off-road/private use only. Use at your own risk. This project is not affiliated with Engwe.
Contributing
Pull requests welcome! Especially interested in:
- Testing on other Engwe models (P275 SE, M20, etc.)
- Cruise control command (
0128) - Motor current limit (
2A66) - Android version of menubar app
License
MIT
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.