canbus-mcp
A CAN bus reverse engineering MCP server for Claude Code that lets Claude read, analyze, and map CAN bus messages from automotive and motorsport ECUs directly from the conversation.
README
canbus-mcp
A CAN bus reverse engineering MCP server for Claude Code. Lets Claude read, analyze, and map CAN bus messages from automotive and motorsport ECUs directly from the conversation.
Built around the PEAK PCAN-USB adapter. Designed for reverse engineering unknown CAN buses without DBC files — though DBC import/export is supported.
Hardware
- PEAK PCAN-USB (IPEH-002021)
Requirements
- Python 3.11+
- macOS or Windows
- PEAK PCBUSB driver (macOS) or PEAK Windows driver
Installation
1. Install the PEAK driver
macOS:
PEAK does not provide a native macOS driver. Use the open-source PCBUSB library from mac-can.com:
- Download the PCBUSB package
- Run the installer — if
install.shdoes nothing,/usr/local/liblikely doesn't exist on your machine. Fix it manually:
sudo mkdir -p /usr/local/lib
sudo cp ~/Downloads/PCBUSB*/libPCBUSB.0.13.dylib /usr/local/lib/
sudo ln -sf /usr/local/lib/libPCBUSB.0.13.dylib /usr/local/lib/libPCBUSB.0.dylib
sudo ln -sf /usr/local/lib/libPCBUSB.0.dylib /usr/local/lib/libPCBUSB.dylib
Windows:
Download and install the PEAK driver from peak-system.com. No additional steps needed.
2. Install canbus-mcp
git clone https://github.com/YOUR_USERNAME/canbus-mcp
cd canbus-mcp
python3 -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -e .
3. Add to Claude Code
claude mcp add canbus-re -s user \
-e DYLD_LIBRARY_PATH=/usr/local/lib \
-- /absolute/path/to/canbus-mcp/.venv/bin/python -m canbus_mcp.server
Replace /absolute/path/to/canbus-mcp with the actual path where you cloned the repo.
Windows: Use the full path to .venv\Scripts\python.exe and omit the DYLD_LIBRARY_PATH env var.
Restart Claude Code and run /mcp to confirm canbus-re appears as connected.
Usage
Start a Claude Code session and describe what you want to reverse engineer. The recommended workflow:
detect_baudrate_auto() # auto-sniff bus speed from live traffic
connect() # open the interface
capture(duration=10) # record frames
get_traffic_summary() # see all message IDs, frequencies, DLCs
take_snapshot('idle') # baseline snapshot
# trigger a physical event (press throttle, turn wheel, etc.)
capture(duration=5)
take_snapshot('event')
compare_snapshots('idle', 'event') # find what changed
analyze_message(0x1A0) # byte-level breakdown of a specific ID
track_signal(0x1A0, 2, 2) # watch bytes [2:4] change over time
define_signal('engine_rpm', 0x1A0, 16, 16, scale=0.25, unit='rpm')
export_dbc('/path/output.dbc')
OBD2
For standard OBD2 vehicles (500k or 250k baud, 11-bit IDs):
check_obd2_support() # check which PIDs are available
read_obd2_pids() # query RPM, speed, throttle, temps, etc.
Tools
| Tool | Description |
|---|---|
list_interfaces |
Show available hardware and current connection |
detect_baudrate_auto |
Auto-detect baud by listening for valid frames |
connect |
Open CAN interface |
disconnect |
Close interface |
capture |
Record frames for N seconds |
take_snapshot |
Save capture buffer as named snapshot |
compare_snapshots |
Diff two snapshots to find changed bytes |
get_traffic_summary |
All IDs: frequency, DLC, count |
analyze_message |
Per-byte entropy, counter/checksum detection |
track_signal |
Extract a byte range's values over time |
define_signal |
Name a discovered signal (bit position, scale, offset) |
decode_frame |
Decode raw hex against known signals |
import_dbc |
Load a DBC file |
export_dbc |
Export defined signals as DBC |
check_obd2_support |
Check OBD2 PID support |
read_obd2_pids |
Query all standard OBD2 PIDs |
query_obd2_pid |
Send a raw OBD2 request |
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.