pixelblaze
Enables Claude to create, update, and manage LED patterns on a PixelBlaze controller, including setting brightness and reading device info, through a set of tools that interact with the device's API.
README
PixelBlaze AI
Create and iterate on LED patterns for a PixelBlaze controller using Claude Code. An MCP server connects Claude directly to your PixelBlaze so it can create, update, and manage patterns on the device.
The included patterns and design guidance (CLAUDE.md) are built for a 2D grid mapped as rows and columns — specifically an edge-lit acrylic depth display where each row is a separate physical layer. The Y axis represents discrete layers (not a smooth gradient), so patterns are designed to create strong visual contrast and coordinated animation across layers while running smooth effects within each layer.
This approach works well for any 2D PixelBlaze setup where one axis has significantly fewer pixels than the other (e.g. 8 rows x 32 columns), and where per-row visual differentiation matters more than smooth vertical blending.
Note: This project assumes the PixelBlaze's Mapper has already been configured correctly for your hardware. The mapper defines the 2D pixel map that tells PixelBlaze which physical LED corresponds to which (x, y) coordinate — patterns depend on this being set up before they can work properly.
Adapting for your hardware
The patterns auto-detect the grid dimensions at startup using mapPixels(), so they adapt to different sizes without code changes. To adapt the design guidance for a different display:
- Edit
CLAUDE.mdto describe your hardware — pixel layout, how light works physically, what effects look good and which don't - Update the pixel map on your PixelBlaze to match your physical layout
- Ask Claude to help you adapt — for example:
"I have a 16x16 LED matrix mounted flat on a wall. Both axes have equal resolution and smooth blending. Update CLAUDE.md to reflect this — we don't need to treat Y as discrete layers, and effects should work as true 2D patterns."
"I have a cylindrical display with 12 rings of 30 LEDs each. X wraps around the cylinder and Y is the vertical axis. Update CLAUDE.md so patterns account for the X-axis wrapping and create effects that look good on a cylinder."
What's in the box
src/pixelblaze_mcp/-- MCP server that exposes PixelBlaze controls (create/update/delete patterns, set brightness, read device info, etc.)docs/pixelblaze/-- Cached PixelBlaze language reference, available to Claude via thedocs_get_api_referencetoolpatterns/-- Pattern JS files with documentation headers, tracked in gitCLAUDE.md-- Project instructions that Claude reads automatically (hardware context, design principles, coding conventions, pattern templates)
Setup
Prerequisites
- PixelBlaze on your local network
- Claude Code CLI installed
- uv (Python package manager)
- Python 3.11+
1. Clone and install dependencies
git clone <repo-url>
cd pixelblaze-ai
uv sync
2. Create the MCP config
Create .mcp.json in the project root (this file is gitignored). This tells Claude Code how to connect to your PixelBlaze:
{
"mcpServers": {
"pixelblaze": {
"type": "stdio",
"command": "uv",
"args": [
"--directory",
"/path/to/pixelblaze-ai",
"run",
"python",
"-m",
"pixelblaze_mcp.server"
],
"env": {
"PIXELBLAZE_HOST": "192.168.2.97"
}
}
}
}
Replace /path/to/pixelblaze-ai with the absolute path to this directory, and 192.168.2.97 with your PixelBlaze's IP.
3. Start Claude Code
cd pixelblaze-ai
claude
Claude will automatically pick up CLAUDE.md for project context and .mcp.json for the PixelBlaze connection.
Usage
Once Claude Code is running, you can ask it to create and modify patterns conversationally. Claude has access to the PixelBlaze API reference and can deploy patterns directly to the device.
Sample prompts
Creating patterns:
- "Create a new pattern that looks like rain falling through the layers"
- "Make a bouncing ball effect with configurable speed and size"
- "Create an aurora effect using Perlin noise"
Modifying patterns:
- "The heartbeat effect has too much dark time between beats, make it faster"
- "Make the fire pattern symmetric -- both edges should be white-hot"
- "Add more randomization to the comet so it doesn't repeat the same path"
Adjusting parameters:
- "Increase the brightness of the fireflies effect"
- "The aurora layers look too similar -- add more color contrast between them"
- "Make the glitch effect more chaotic with panel blackouts and freezes"
Working with the display:
- "List all patterns on the device"
- "Switch to the plasma pattern"
- "Set the display brightness to 50%"
Available MCP tools
Claude has access to these PixelBlaze tools:
| Tool | Description |
|---|---|
pixelblaze_create_pattern |
Create and activate a new pattern |
pixelblaze_update_pattern |
Update an existing pattern's code |
pixelblaze_delete_pattern |
Remove a pattern from the device |
pixelblaze_list_patterns |
List all patterns on the device |
pixelblaze_set_active_pattern |
Switch to a specific pattern |
pixelblaze_get_active_pattern |
See which pattern is running |
pixelblaze_get_pattern_code |
Read a pattern's source code |
pixelblaze_set_brightness |
Set display brightness (0-1) |
pixelblaze_get_controls |
Read current slider/control values |
pixelblaze_set_control |
Set a slider/control value |
pixelblaze_get_device_info |
Get device status and config |
docs_get_api_reference |
Get the PixelBlaze language reference |
Project structure
pixelblaze-ai/
.mcp.json # MCP server config (gitignored)
CLAUDE.md # Project instructions for Claude
pyproject.toml # Python project config
src/pixelblaze_mcp/ # MCP server source
docs/pixelblaze/ # Cached API reference
patterns/ # Pattern JS files (deployed to device)
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.