← Back to directory
B

Brandsystem MCP

Community
Turn existing brand guidelines into AI-usable portable brand runtime.
Category
Dev Tools #95 of 438
Stars
★ 15 Popular
Transport
stdio (local process)
Runtime
Node.js 20.18.1+
Credentials
Optional API key
License
MIT
Last commit
Tools
33
62FMRS · C

Brandsystem MCP provides a comprehensive suite of brand management tools enabling AI agents to follow brand guidelines. It supports extracting brand elements from websites, Figma, etc., generates design tokens and brand runtime, and provides checking, auditing, and compliance capabilities. It is particularly suitable for design and development teams aiming to integrate brand systems into AI workflows. However, be aware that extraction accuracy depends on algorithms and may require human intervention, and there is some configuration complexity.

Strongest · Setup experience 14/20 Weakest · Reliability 10/20

Reliability
10/20
Security and permissions
12/20
Maintenance
12/20
Documentation
14/20
Setup experience
14/20
Why each score
Reliability 10/20
README claims CI badge, a structured test directory (9 lib tests, 2 tool tests, server smoke test), and an eval harness, but a static review cannot verify the tests actually cover key paths; manifest and npm versions align (0.16.1), and stdio transport, Node >= 20.18.1, 5K/50K truncation, and error paths (empty extraction, Figma's external-MCP dependency) are documented. Under static calibration this stays at or below 12; scored 10 because claimed tests and error handling are plausible but not verifiably executed.
Security and permissions 12/20
No red-line behavior: local-first, network limited to user-directed fetches, optional Brandcode connector, and optional brand_feedback; connector credentials in gitignored .brand/brandcode-auth.; hosted side is read/append-only with bearer-key auth and explicit capture scope; install is dry-run by default, deep-merges, and backs up. Deductions: credential permission details not shown, hosted service is pre-release/approved-only, and data-boundary claims rest on README self-description rather than inspectable code.
Maintenance 12/20
MIT license, SECURITY.md private reporting channel, CI badge, npm version advanced to 0.16.1, and specs/eval directories indicate sustained engineering. Deductions: no static evidence of release cadence, dependency-update policy, or issue responsiveness; 14 open issues against 10 stars with no stated triage; the hosted half is not public.
Documentation 14/20
Layered docs: per-client setup, CLI reference, tool tables, .brand file details, confidence/source-precedence model, troubleshooting (empty extraction, Figma, logo, 5K limit, startup failures), and eval methodology with disclosed limitations (static-CSS parsing limits, pre-release hosting). Deductions: cost/quota information absent, tool parameter details deferred to code, and Claude Design integration claims lack independent evidence.
Setup experience 14/20
Clear install path: npx install supports multiple clients (Claude Code/Cursor/Windsurf/Cline/Codex/Claude Desktop) with .mcp. examples, dry-run default, deep-merge with backup, plus doctor/inspect diagnostics. Deductions: static rules prevent verifying the install actually works; Figma flow requires a separate MCP server and headless Chrome is an implicit dependency, adding setup complexity.

Static review · not runListed 2026-09-08

Read the FMRS scoring method →

Fit and risk

What it can accessReads local filesWrites / deletes local filesUses the networkControls a browser

Best for

  • Design teams needing consistent brand identity in AI-driven content.
  • Developers who want to automate brand guideline extraction and token compilation.
  • Teams using Claude Code, Cursor, and similar tools.
  • Marketing professionals and content creators managing brand assets.

Not for

  • General AI tasks that don't need brand management.
  • Users who only want a hosted brand service (might be better with @brandcode/mcp).
  • Those relying on Figma direct integration without extra Figma MCP server.
  • Users without Node.js 20+ or who prefer not to run local servers.

Required permissions

  • Reads the websites you specify for extractions (network access).
  • Reads local filesystem and writes to the .brand directory (within your project dir).
  • Accesses authentication credentials when connecting hosted Brandcode Studio (stored in .brand/brandcode-auth.json).
  • Sends optional feedback to the brand team.

Risks and side effects

  • Network communication with remote sites, possible copyright/privacy concerns.
  • Extracted data is algorithmically inferred, may have inaccuracies, needs human review.
  • Figma extraction requires an additional Figma MCP, adding configuration complexity.
  • Default core profile only exposes 12 tools; full functionality requires opting into full profile.

Setup

Before you start

Runtime:Node.js 20.18.1+

ANTHROPIC_API_KEY optionalsecret Only needed to run the repo's model-dependent evaluation tier (first-tool selection); obtain from the Anthropic console. Not required for normal server use.
Other optional settings (2)
BRANDSYSTEM_PROFILE optional Optional; set to full to register all tools instead of the default 12-tool Core profile. Can also be passed as --profile=full.
BRANDCODE_MCP_SMOKE_URL optional Maintainer-only target URL for the hosted deployment smoke harness (npm run smoke:hosted-mcp); not needed by end users.
  1. Create .mcp.json in project root (Claude Code) or .cursor/mcp.json (Cursor) or use CLI installer.
{
  "mcpServers": {
    "brandsystem": {
      "command": "npx",
      "args": ["-y", "@brandsystem/mcp"]
    }
  }
}
  1. For Codex, Cline, etc., use npx @brandsystem/mcp install --client <client> --write to auto-configure.
  2. Ensure Node.js >= 20.18.1.
claude_desktop_config.json
{
  "mcpServers": {
    "brandsystem": {
      "command": "npx",
      "args": [
        "-y",
        "@brandsystem/mcp"
      ]
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "brandsystem": {
      "command": "npx",
      "args": [
        "-y",
        "@brandsystem/mcp"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add brandsystem -- npx -y @brandsystem/mcp

Run it in a terminal; replace any <…> placeholders with your own values first.

Check that it works

After adding npx -y @brandsystem/mcp to your client config, brand_start and brand_runtime should appear in the tool list; calling brand_status without a .brand/ directory should return the getting-started guide, proving the connection works.

Troubleshooting

  1. If no .brand directory found, run brand_start first.
  2. If extraction yields no colors/fonts, site may be dynamic: try brand_extract_visual or brand_extract_site.
  3. If Figma extraction fails, ensure a separate Figma MCP server is connected.
  4. If logo not detected, use brand_set_logo to add manually.
  5. If server won't start, check Node.js version (>=20.18.1) and run `npx @brandsystem/mcp doctor`.

Things to try

Once connected, you can ask your AI assistant things like:

  • Run brand_start with client_name="Acme Corp", website_url="https://acme.com", and mode="auto" to create a brand system
  • Run brand_extract_visual on this JS-rendered site, then brand_compile to generate the design tokens
  • Run brand_write for a social graphic about the Q3 product launch using my brand guidelines
  • Check this draft with brand_check_compliance and generate a shareable HTML report with brand_report

Tools 33

brand_start writes
Begin here. Creates a brand system from a website URL in under 60 seconds, with auto mode for one-call setup.
brand_status read-only
Check progress, get next steps, or see a getting-started guide if no brand exists yet.
brand_extract_web writes
Extract logo (SVG/PNG), colors, and fonts from any website URL.
brand_extract_visual writes
Screenshot the rendered page in headless Chrome and extract computed colors, fonts, and visual context from JS-heavy sites.
brand_extract_site writes
Discover representative pages, render them across desktop and mobile, capture screenshots, sample multiple components, and persist extraction-evidence.json.
brand_extract_figma writes
Extract from Figma design files (higher accuracy). Two-phase: plan then ingest.
brand_set_logo writes
Add/replace logo via SVG markup, URL, or data URI.
brand_compile writes
Generate DTCG design tokens, brand runtime contract, and interaction policy from extracted data.
Show 25 more tools
brand_clarify writes
Resolve ambiguous brand values interactively (color roles, font confirmations).
brand_audit read-only
Validate .brand/ directory for completeness and correctness.
brand_report writes
Generate portable HTML brand report. Upload to any AI chat as instant guidelines.
brand_init writes
Low-level directory scaffolding. Prefer brand_start instead.
brand_deepen_identity writes
Define composition rules, patterns, illustration style, and anti-patterns (6 interview sections).
brand_ingest_assets writes
Scan and catalog brand assets with MANIFEST.yaml metadata.
brand_preflight read-only
Check HTML/CSS against brand rules -- catches off-brand colors, wrong fonts, anti-pattern violations.
brand_extract_messaging writes
Audit existing website voice -- fingerprint, vocabulary, claims, AI-isms, gaps.
brand_compile_messaging writes
Define perspective, voice codex (tone, vocabulary, AI-ism detection), and brand story.
brand_build_personas writes
Build buyer personas through a 7-question guided interview.
brand_build_journey writes
Define buyer journey stages (ships with 4 proven defaults).
brand_build_themes writes
Define editorial content themes balanced across awareness, engagement, and conversion.
brand_build_matrix writes
Generate messaging variants for every persona x journey stage combination.
brand_audit_content read-only
Score content against brand rules (0-100) across multiple dimensions.
brand_check_compliance read-only
Quick pass/fail compliance gate before publishing.
brand_audit_drift read-only
Detect systematic brand drift across multiple pieces of content.
brand_runtime read-only
Read the compiled brand runtime contract (single-document brand context for AI agents).
brand_context read-only
Select a task-scoped slice of the runtime deterministically (task_type → sections, audience → persona match, compact budget). Returns matched selectors and explicit no-match.
brand_write read-only
Load full brand context (visual + voice + strategy) for content generation.
brand_export writes
Generate portable brand files for Chat, Code, team sharing, or email.
brand_feedback destructive
Report bugs, friction, or feature ideas to the brandsystem team.
brand_brandcode_connect writes
Connect a local .brand/ directory to a hosted Brandcode Studio brand and pull the current package.
brand_brandcode_sync writes
Pull updates from a previously connected hosted brand using sync-token-aware delta behavior.
brand_brandcode_status read-only
Inspect the current Brandcode Studio connection, sync history, and local package summary.
brand_brandcode_live writes
Toggle connected read tools to refresh from the hosted runtime within a short cache TTL.

Use cases

Rapidly extract brand colors, fonts, and logos from a website.
Import brand tokens from Figma design libraries.
Define brand voice and anti-patterns for AI content generation.
Check content compliance against brand rules before publishing.
Create portable brand reports for sharing with teams and AI tools.
Connect to Brandcode Studio for syncing and single source of truth.

Supported clients

Claude Desktop
Claude Code
Cursor
Windsurf
Cline
Codex

Listed from the project's documentation, not tested by this site.

Overview

Brandsystem MCP server is designed to convert existing brand materials, such as websites, PDFs, Figma files, into a portable .brand runtime that AI agents can understand and use. It provides a complete suite of tools covering the four stages of brand adoption: Core Identity, Visual Identity, Messaging, and Content Strategy. It generates structured brand data including design tokens, governance rules, provenance, and compliance checks, enabling AI to produce on-brand content consistently. The server is local-first, no account required by default, runs over stdio, and is compatible with multiple development tools.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 63.3k · Tools 2 Compare with this →

Source revision eded428fa0b9 Data synced 2026-10-19 Read the FMRS scoring method