Brandsystem MCP — Brand Guidelines for AI
Turn a website, PDF or Figma library into a .brand runtime AI tools can follow
- Platforms
- CLI, macOS, Windows
- Runs
- Locally
- Version
- 0.18.0
- Listing updated
- 4 days ago
Similar MCP servers
About Brandsystem MCP — Brand Guidelines for AI
Brandsystem MCP — Brand Guidelines for AI is an MCP server for design and media. It runs locally on your machine and is started by your AI client.
Brandsystem MCP starts from the argument that AI output is often correct but generic because brand guidelines sit in PDFs, Figma files and people's heads. It reads what you already have, whether a website, PDF guide, Figma library or local files, and compiles a portable .brand directory with design tokens, voice rules, anti-patterns, proof-point status and compliance checks, plus a brand-runtime.json file that agents load at generation time.
The default Core profile exposes 12 tools covering an adopt, use and check loop, and the setup command is a dry run until you add the write flag. Claude Desktop users can install a one-click bundle. The project describes a second, hosted MCP that serves the same runtime at mcp.brandcode.studio for teams that want authenticated live reads. This package is the authoring half, running locally on your machine, and the README says no account is required.
It works with Claude Code, Claude Desktop, Cursor, Windsurf and Cline over stdio, while ChatGPT and other remote clients can use the runtime as an uploaded file. The README says brand prompts shrink from several hundred tokens of inline context to just the delta.
Install
Run in a terminal
claude mcp add brandsystem-mcp -- npx -y @brandsystem/mcpAdd to claude_desktop_config.json
{
"mcpServers": {
"brandsystem-mcp": {
"command": "npx",
"args": [
"-y",
"@brandsystem/mcp"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"brandsystem-mcp": {
"command": "npx",
"args": [
"-y",
"@brandsystem/mcp"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"brandsystem-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@brandsystem/mcp"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"brandsystem-mcp": {
"command": "npx",
"args": [
"-y",
"@brandsystem/mcp"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"brandsystem-mcp": {
"command": "npx",
"args": [
"-y",
"@brandsystem/mcp"
]
}
}
}Using several servers? Build one config file with the MCP config generator.
What it does
- Brand extraction
- Builds brand data from a website, PDF, Figma library or local files.
- .brand runtime
- Compiles tokens, voice rules and governance into one portable directory.
- Compliance checks
- Audits output against the brand rules with provenance.
- Local-first
- Runs on your machine without an account.
- Easy client setup
- An install command configures several clients, with a dry run by default.
- Hosted companion
- A separate hosted MCP serves the runtime to teams.
Good for
- Make AI-generated copy follow your voice rules
- Turn a PDF brand guide into machine-readable tokens
- Check a draft for brand compliance
- Give Cursor or Claude Code your design tokens
- Share one brand runtime across tools and repos
Brandsystem MCP — Brand Guidelines for AI FAQ
Do I need an account?
No. The README says it is local-first with no account required.
What can it read?
A website, PDF guide, Figma library, local files or a Brandcode Studio brand.
How many tools does it expose?
The default Core profile exposes 12 tools.
Which clients are supported?
Claude Code, Claude Desktop, Cursor, Windsurf and Cline directly, and ChatGPT via an uploaded runtime file.
How do I add Brandsystem MCP — Brand Guidelines for AI to Claude Code?
Run this command in your terminal: claude mcp add brandsystem-mcp -- npx -y @brandsystem/mcp
What are the best alternatives to Brandsystem MCP — Brand Guidelines for AI?
The closest alternatives to Brandsystem MCP — Brand Guidelines for AI are Brandpilot MCP, Rainbrand, Dante IX — Autonomous Brand & GEO Agency MCP, Ads AI Creator MCP and Marketing & Brand Context by Marcora.
Where this comes from
- Description: written from the project's README and the project's website.
- Install steps: from the official MCP Registry entry.