Brandize

Registre officiel Recherche et web

AI logo generation, brand color palettes, SEO meta tags, JSON-LD schema markup, and blog search.

Installation

Ajoutez ceci à la configuration MCP de votre client (Claude Desktop : claude_desktop_config.json ; Cursor : .cursor/mcp.json ; VS Code et les autres utilisent la même forme « mcpServers »).

Remote (streamable-http)
Claude Code
claude mcp add --transport http brandize 'https://brandize.me/api/mcp'
Configuration JSON (Claude Desktop, Cursor, VS Code)
{
    "mcpServers": {
        "brandize": {
            "type": "http",
            "url": "https://brandize.me/api/mcp"
        }
    }
}

Documentation

Tirée du README du projet sur GitHub, actualisée chaque semaine. Les liens ouvrent GitHub.

Sommaire
  1. Brandize MCP server
  2. Connect
  3. Tools
  4. Registry listing

Brandize MCP server

Hosted MCP server for Brandize, an AI logo generator. It generates logos and full brand kits — lockups, favicons, social assets, print collateral — plus brand color palettes, SEO meta tags, and JSON-LD schema markup, and searches the Brandize design guides.

There is nothing to install. The server is remote and runs over streamable HTTP:

Endpointhttps://brandize.me/api/mcp
Transportstreamable-HTTP
Authnone
Discovery manifesthttps://brandize.me/.well-known/mcp
Docshttps://brandize.me/mcp
Machine-readable site docshttps://brandize.me/llms.txt

This repository holds the documentation and the Claude Code and Codex plugin (plugins/brandize) — the server itself is part of the Brandize application. Issues and questions about the MCP surface belong here.

Connect

Claude Code — plugin

/plugin marketplace add minim-l/brandize-mcp
/plugin install brandize@brandize

Codex — plugin

codex plugin marketplace add minim-l/brandize-mcp
codex plugin add brandize@brandize

Or install brandize from /plugins inside Codex.

Claude Code — CLI

claude mcp add --transport http brandize https://brandize.me/api/mcp

Clients that take a JSON config (Claude Desktop, Cursor, Windsurf)

{
  "mcpServers": {
    "brandize": {
      "type": "http",
      "url": "https://brandize.me/api/mcp"
    }
  }
}

MCP Inspector

npx @modelcontextprotocol/inspector
# transport: Streamable HTTP → https://brandize.me/api/mcp

Call echo first to confirm the endpoint is reachable.

Tools

Free, no account, rate-limited:

ToolWhat it does
generate_color_paletteFive-color brand palette from a base hex color and a harmony rule. Returns CSS custom properties, a Tailwind snippet, or plain hex.
generate_meta_tagsAn HTML <head> block — title, description, canonical, robots, Open Graph, Twitter Card.
generate_schema_markupA JSON-LD <script> block for LocalBusiness, Organization, WebSite, Article, FAQPage, or Product.
search_blogKeyword search over the Brandize logo design and branding guides.
get_pricing_tiersCurrent service tiers, prices, and included features. Read-only.
generate_logo_variationsAlternate style directions for an existing logo job, within that job's free preview credits.
echoHealth check. Returns the message unchanged.

Paid — each mints a Stripe checkout link the user opens and pays in a browser. No card is stored and no call charges anyone on its own:

ToolWhat it does
generate_logoGenerates a watermarked, low-resolution preview from a design brief, plus a checkout link for the tier the user picked.
buy_logo_variationBuys one of the alternates from generate_logo_variations instead of the original design.
get_logo_resultPolls a job by its jobToken. Once payment settles, returns a download URL for the tier's full deliverable.

generate_logo asks the calling agent to collect a real brief — tier, style, colors, layout, surface, industry — and to take the user's explicit acceptance of the Terms and Privacy Policy before it will generate.

Tiers

The tier argument to generate_logo picks what gets delivered. Call get_pricing_tiers for the current prices and feature lists — they are set server-side, not hardcoded here.

TierDeliverable
PREMIUMFull-resolution PNG, vector SVG, commercial license.
COMPLETEAdds alternate logo variations, assembled after payment.
BRAND_STARTERA brand kit: icon, horizontal, stacked, monochrome and knockout lockups, the full favicon package, a five-color palette, a font pairing, a print-ready business card PDF, and an HTML email signature.
BRAND_KITEverything in Brand Starter, plus a social media kit across five platforms, a brand guidelines PDF, a letterhead template, an Open Graph image, and iOS/Android/PWA app icons.

The brand-kit tiers have fixed deliverables — generate_logo_variations refuses on a purchased one. Only COMPLETE accepts more alternates after payment.

A paid run, end to end
  1. generate_logo → watermarked preview, a jobToken, a checkoutUrl, and the higher tiers on offer.
  2. The user opens checkoutUrl and pays.
  3. get_logo_result with that jobToken, polled every ~5 seconds — payment confirmation is not instant — until the status is paid.
  4. Download the returned asset URL.

A refunded purchase reverts the job: get_logo_result reports pending_payment again and the asset URL stops resolving.

Registry listing

The official MCP Registry entry is me.brandize/brandize, published from server.json in the Brandize application repo and verified against the brandize.me domain.

Questions fréquentes

Qu'est-ce que Brandize ?

AI logo generation, brand color palettes, SEO meta tags, JSON-LD schema markup, and blog search.

Comment installer Brandize ?

Il fonctionne comme serveur distant. Dans Claude Code : claude mcp add --transport http brandize 'https://brandize.me/api/mcp'. Pour les autres clients, collez la configuration JSON de cette page dans leurs réglages MCP.

Faut-il une clé API pour Brandize ?

Son entrée au registre n'indique aucune clé ni secret obligatoire. Certaines fonctions peuvent tout de même demander un compte auprès du service auquel il se connecte.

Brandize est-il open source ?

Oui. Son code est sur GitHub sous licence MIT.

Qui maintient Brandize ?

Il est publié par minim-l. AI.tn le référence sans le maintenir ni le recommander.

Issu du registre MCP officiel. Être listé n'est pas une recommandation : vérifiez le code et les permissions d'un serveur avant de lui donner accès à vos données.