saglitzdesign mcp

SaglitzDesign MCP

An expert design & marketing brain for your AI coding agent.

A Model Context Protocol server that gives Claude, Cursor, and any MCP client expert‑level guidance on web, iOS, Android and macOS design — plus the UX, copywriting, SEO, GEO and marketing knowledge that makes a product actually convert.

96 curated knowledge documents · 34 tools · 8 build/review/port workflows · MCP resources with id autocomplete · a one‑call design‑system builder · real token/color/type/elevation/motion/a11y generators · design & UX‑copy linters · production component recipes · phased roadmaps · real‑world visual examples

Why · What's inside · Tools · Resources · Install · Usage · Your own rules · Changelog · License

---

Why

LLMs are confidently wrong about design. They reach for defaults, invent outdated specs, and give generic "make it clean and modern" advice. SaglitzDesign replaces that with a curated, sourced, prescriptive knowledge base your agent can query on demand — the kind of guidance you'd get from a senior product designer, a conversion copywriter, and a technical SEO all at once.

Ask your agent to design a paywall, audit a landing page, or plan an iOS app, and it pulls concrete rules ("44pt minimum touch target", "one primary button per screen", "LCP ≤ 2.5s"), real patterns from top apps, and a phased roadmap — instead of guessing.

• Runtime‑independent. The server reads only local files. No external API, no account, nothing to configure. It just works, offline.
• Local by design. It speaks MCP over stdio only and runs as a child process of your MCP client. There is no HTTP or SSE transport, so it cannot be hosted remotely or reached over a network — which is also why your prompts and code never leave your machine.
• Prescriptive, not vague. Every doc is written as rules an agent applies verbatim — numbers, thresholds, do/don't lists, anti‑patterns.
• Grounded. Design‑language specs from official sources; patterns studied from real top apps and sites; classics distilled from the actual books.

What's inside

96 knowledge documents across 11 categories:

Category	Coverage
🎨 Design languages	Material 3 & M3 Expressive · Apple HIG + Liquid Glass (iOS 26) · deep iOS, Android (Android 16 / M3 Expressive) and macOS app‑design guides · Apple accessibility (Dynamic Type, VoiceOver, hit regions) and Apple shipping readiness (Info.plist purpose strings, entitlements, app icon sets, orientation/multitasking restrictions) · Apple Intelligence design (AI features: Writing Tools, App Intents, on‑device Foundation Models) · visionOS / spatial design (Vision Pro) · Fluent 2 · 2026 web trends · design tokens & theming (W3C DTCG) · Apple WWDC design principles (fluid interfaces)
🧩 Components	Buttons (hierarchy, sizing, states, labels) · forms & inputs · navigation · cards / lists / modals / sheets / empty states
🧠 UX	Nielsen heuristics & behavioral laws · accessibility (WCAG 2.2) · typography · color & dark mode · spacing & grids · motion · mobile UX · conversion / CRO · data visualization · information architecture · i18n / localization (RTL) · AI product UX (chat, streaming, agentic) · onboarding & permission priming
✨ Craft	Expert polish standards · typographic craft · animation craft (easing, springs, interruptibility) · UX writing & cognitive load · 0–40 critique rubric · clean/minimal app design · design‑engineering (semantic HTML, CSS architecture, tokens‑in‑code) · ethical design (avoiding dark patterns) · iconography (choosing & using an icon system) · the AI‑default aesthetic (the stock gradient, font, card chrome and copy generated interfaces reach for, cited to each system's own docs)
📚 Books	Distilled classics — design: Norman, Krug, Refactoring UI, psychology of design, grid/typography, interaction design (Cooper/Tidwell), emotional design (Walter/Norman) · marketing: Cialdini, Positioning, StoryBrand + Ogilvy, Hooked
🗺️ Process	Product‑design & marketing‑website roadmaps · design‑systems methodology (Atomic Design, component API, governance) · design handoff (Figma Dev Mode, Code Connect, design↔dev)
📣 Marketing	Branding & identity · email marketing · HTML email development (Outlook, dark mode, bulletproof) · ad creative · paywall benchmarks (RevenueCat 2026) · growth frameworks (loops/AARRR/PLG) · pricing strategy · analytics & experimentation · value proposition & JTBD · content & distribution (topic clusters, community, referral) · App Store Optimization (ASO)
🔎 SEO	Technical SEO (Core Web Vitals) · on‑page & E‑E‑A‑T · SEO for designers
🤖 GEO	Generative Engine Optimization — visibility in ChatGPT / Perplexity / AI Overviews, llms.txt, citation tactics
🔐 Security	Web security headers & CSP — strict nonce/hash policies, strict-dynamic, Trusted Types, HSTS, SRI, cross-origin isolation, and the superseded headers auditors still ask for. Every claim sourced to a spec or vendor doc, re-verified on a 90-day clock.
🖼️ Patterns & examples	Real‑world patterns studied from top apps & sites (incl. e‑commerce & checkout and fintech / trust flows), plus a curated library of real‑world example screens

Which documents' sources are enforced, and which are not. Of the 96 documents, 11 have their sources: checked by the test suite against a tiered allowlist — the six Apple design‑language guides (apple-hig-liquid-glass, ios-app-design, macos-app-design, apple-accessibility, apple-shipping-readiness, wwdc-design-principles) and the five security documents. The other 85 are curated but not yet checked, and extending the assertion to them is its own piece of work rather than a formality: measured today, 66 of the 85 would fail on 282 sources between them — 227 URL citations across 133 distinct hosts, plus 55 that are not URLs at all and so have no host to check. (The 227 and the 55 are the two halves of the 282, not two figures to add.) Note that apple-intelligence-design and visionos-spatial-design are Apple‑topic documents that sit outside the enforced set. get_design_doc prints which side of this line a document is on, next to its sources, so you never have to come back here to find out.

Workflows (/ prompts) — "build me a…"

Beyond answering questions, SaglitzDesign ships prompts that orchestrate an entire build end‑to‑end. In Claude Code they appear in the / menu under a name that depends on how you installed it. Installed as the Claude Code plugin they are /saglitzdesign:build_landing_page and so on — plugin commands, one generated file per workflow in commands/. With the server installed on its own they are /mcp__saglitzdesign__build_landing_page, which the menu labels saglitzdesign:build_landing_page (MCP). That label is typeable too, (MCP) suffix and all — it is just a clumsier way to reach the same prompt. What does not work is that short form with the suffix dropped: Claude Code hands the bare server:prompt alias to first‑party Anthropic connectors only — the URL has to be https, on api.anthropic.com, under /v1/design/ — and no other server gets one, stdio or remote. (The plugin's /saglitzdesign:build_landing_page above only looks like that alias — it is a plugin command namespace, a different mechanism.)

Invoke one and the agent runs the full method — roadmap → positioning & copy → generates the design system (color, type, layout, elevation, tokens) → real examples → writes the actual code from the component recipes → runs the deterministic verify gate (design_lint, audit_accessibility, audit_design_system, audit_ux_copy) → opens it in a browser, screenshots, scores it against the critique rubric, and iterates until it passes.

Workflow	What it does
build_landing_page	Designs & builds a conversion‑focused landing page, copy‑first, with a visual critique loop.
build_website	Builds a multi‑page marketing site — positioning, IA, SEO/GEO, shared design system.
build_mobile_app_ui	Builds iOS or Android screens on the correct platform baseline (HIG/Liquid Glass or Material 3).
critique_screenshot	Measures the screenshot, then critiques it against the fixed 0–40 rubric — cites real ratios and colour counts, specific elements, no padding.
review_paywall	Scores a paywall / subscription onboarding against real RevenueCat 2026 conversion benchmarks.
design_review	Audits an existing site/app — runs the deterministic auditors first, so findings lead with measured numbers, then the checklists and the 0–40 rubric, ranked by severity.
redesign	Improves an existing UI (bolder / quieter / higher‑converting) using the craft standards, with a measured before→after (consistency score, critique score, lint findings, contrast failures).
port_to_platform	Takes an existing UI to another platform (iOS ↔ Android ↔ macOS ↔ web) surface by surface — porting the intent and IA, never the components.

Just type, e.g., /saglitzdesign:build_landing_page a SaaS invoicing tool for freelancers — everything after the command name becomes the brief, and the workflow asks for anything missing, then builds it. The /mcp__saglitzdesign__… form takes only the first whitespace‑separated word as its brief, so use the plugin command whenever the brief is a phrase. These are yours to type: every command carries disable-model-invocation, so the agent never starts one on its own. Asked in prose it reaches for the skills instead, and they cover much of the same ground — design review, landing‑page conversion, Apple platforms — as guidance rather than as this orchestrated build. The saglitzdesign skill's own trigger vocabulary names paywalls, so prose about one does reach a skill; what no skill carries is review_paywall's RevenueCat benchmark scoring, which runs only when you type it.

The visual critique loop uses whatever browser tool is connected (Claude in Chrome, Playwright, or chrome‑devtools MCP) to see and refine its own output. Without one, it reviews the code directly.

Tools

Tool	What it does
create_design_system	The one‑call foundation. Brand color + vibe + platform → a complete, coherent starter: WCAG‑verified color (light+dark), matched fonts, an icon library, a modular type scale, an elevation ramp, ready‑to‑paste tokens, the components to build, and a checklist — all generated to work together.
get_design_roadmap	Start here for process. A phased, expert process for a project type (website, landing page, iOS / Android / macOS app, SaaS web app) — each phase has a goal, exit criteria, and the exact docs to read.
search_design_knowledge	Natural‑language search across everything, returning the most relevant section of the best‑matching docs.
get_design_doc	Fetch any document in full by id.
get_component_guidance	Deep dive on a component or screen (button, form, paywall, hero, pricing…) — specs + real‑world patterns.
get_design_language	Full platform / design‑system references (Material 3, Liquid Glass, iOS/Android/macOS, Fluent 2, web trends, tokens).
compare_design_languages	Building on more than one platform? Side‑by‑side iOS / Android / macOS / web conventions for one surface (navigation, buttons, sheets, motion, forms…), plus the porting rules — and an explicit do NOT port list.
get_design_examples	Curated real‑world examples of a pattern from top apps/sites, with notes on what each does well. (Screenshots are a local‑only asset — see Visual examples.)
design_review_checklist	An assembled audit checklist per project type and focus (UI, UX, accessibility, SEO, GEO, conversion, copywriting).
seo_geo_guide	SEO and GEO guides, optionally narrowed to a topic.
import_design_tokens	Already have a design system? Paste your CSS custom properties, shadcn :root block, DTCG file or theme JSON and get back the roles it names, the roles it leaves undefined, a WCAG check on the pairs it defines, and the whole set re‑emitted as CSS / Tailwind / SwiftUI / Compose / DTCG. The inverse of generate_design_tokens — take a web theme to iOS, or audit an inherited one. Only named tokens are read; JS configs are never evaluated.
generate_design_tokens	Real artifacts, not advice — turns a color/spacing/type spec into CSS variables, Tailwind v4, SwiftUI, Jetpack Compose, and W3C DTCG JSON.
audit_accessibility	Deterministic WCAG 2.2 checks — exact contrast ratios for color pairs + tap‑target sizes per platform, with fixes.
get_component_recipe	Production‑ready, accessible reference code for a component (button, input, modal, toast, card, switch, tabs, empty‑state, list‑row) in react‑tailwind, html‑css, SwiftUI, or Compose — all states, ARIA, keyboard, correct motion. Pass your scales — the neutral, primary and danger ramps generate_color_system returns — and the code comes back in your colours, dark‑mode shades included.
generate_color_system	One brand color → a full palette. A 50–950 tonal scale, a cohesive neutral ramp, status colors (danger / success / warning) harmonised with your brand, and light + dark semantic tokens — every text/UI pair WCAG‑verified and auto‑corrected. Feeds straight into generate_design_tokens.
suggest_font_pairing	Curated, production font pairings for a vibe (SaaS, editorial, bold, native…) — heading + body (+ mono) with paste‑ready CSS stacks, weights, rationale, and a type scale.
fix_contrast	Repairs a failing color pair: computes the nearest accessible color (hue/saturation preserved) that meets your WCAG target — the corrected value, not just a fail report.
suggest_icon_library	Recommends the right icon system for a vibe/platform (Lucide, Phosphor, Solar, SF Symbols, Material Symbols…) — with license, install command, coverage, fit rationale, and usage rules.
generate_type_scale	A modular type scale (base × ratio) → named steps with line‑heights, tracking, and fluid clamp(), as CSS variables + Tailwind.
generate_elevation_system	A cohesive layered box‑shadow ramp (flat→modal) as CSS variables + Tailwind, with dark‑mode guidance.
generate_layout_system	Breakpoints (with what changes at each), container widths, a column grid, an intrinsic auto‑fit grid, container queries, and a fluid section‑rhythm scale — CSS variables + Tailwind v4.
generate_motion	Easing + duration tokens and ready‑to‑paste keyframe animations (fade/slide/scale/spring/shimmer) in CSS, Framer Motion, or SwiftUI — reduced‑motion included.
design_lint	Lints an HTML/CSS/JSX/Tailwind snippet for design & a11y anti‑patterns (hardcoded values, killed focus, missing alt/labels, clickable divs, unlabelled inputs…) with line numbers and fixes. Tag‑aware, so formatting never changes the verdict. Returns markdown plus structured output — findings, a severity summary, and a machine‑readable notVisible list of what it could not check.
measure_screenshot	Measures your actual screen. Give it a PNG and it reports the real palette and colour count, true WCAG contrast ratios for the pairs on screen, density, and structural detections (alignment, rhythm, off‑grid gaps) each with a confidence level — plus a self‑contained HTML report you can open and share. Pure‑Node PNG decoding, no network, no dependencies.
audit_project	Audits a real codebase, not a snippet. Point it at a directory: it walks your design source, lints every file, and scores the whole project for consistency — cross‑file drift being exactly what a single‑file lint cannot see. Findings ranked worst‑file‑first with file:line, plus an explicit list of what it did not look at. Returns markdown plus structured output — findings, a severity summary, a machine‑readable notVisible list, and a scan block reporting how many files and bytes were actually read, which were skipped for size or could not be opened, and whether a cap was hit. A missing or non‑directory path comes back as an error result, not an empty audit.
audit_security	Audits a web project or snippet for the defects that actually ship — missing or weak Content‑Security‑Policy, absent HSTS, unpinned cross‑origin scripts, mixed content, credentials in localStorage, secret‑named NEXT_PUBLIC_/VITE_ variables, unsandboxed third‑party iframes, wildcard postMessage, raw‑HTML sinks with no sanitiser, production source maps and un‑ignored .env files. Header state is inferred from wherever your stack declares it — next.config, vercel.json, netlify.toml, _headers, staticwebapp.config.json, firebase.json, Nuxt routeRules, a Remix headers export, SvelteKit hooks.server.ts / kit.csp, Astro middleware and Next middleware or its Next‑16 rename proxy.ts, new Response(…, { headers }) and new Headers({…}) on Workers/Deno/Bun, a quoted header‑name property in any JSON or object literal, any call whose method is set, setHeader, append or header whatever the object is named (res.set / res.setHeader, headers.set / headers.append, Fastify reply.header, Hono c.header, Koa ctx.set), and <meta http-equiv> — read as text and never evaluated; this makes no network request, and says what it could not see. Returns markdown plus structured output — findings, a severity summary, and a machine‑readable notVisible list of what it could not check. A missing or non‑directory path comes back as an error result, not an empty audit.
audit_seo_geo	Audits the SEO and GEO signals that are actually in your source — a missing <title> and one outside the width a result gives it, the same for the meta description, multiple H1s, skipped heading levels, a missing canonical, one written relative in a self‑contained document, one left pointing at localhost or staging, an hreflang set that never lists the page itself, JSON‑LD that does not parse or declares no @context/@type or declares a type whose rich result Google has retired, missing alt text, robots.txt crawl rules (including the AI crawlers behind ChatGPT, Claude, Perplexity and Google's AI surfaces), a robots.txt naming no sitemap, no llms.txt beside it, and content that only exists once a script has run. Absence is claimed only where it can be proven — a self‑contained HTML document, or a whole directory — and a capped scan downgrades every absence claim to an unconfirmed note. It reads source and measures nothing: no request is made, so no finding is a Core Web Vitals result, an indexing status or a ranking outcome. Returns markdown plus structured output — findings, a severity summary, and a machine‑readable notVisible list of what it could not check.
audit_performance	Audits the performance signals that are actually in your source — a hero image held back by loading="lazy" or contradicting its own fetchpriority, the LCP‑candidate image declaring no fetch priority at all, a hero background declared in CSS or an inline style that the HTML preload scanner never sees, images with no width/height or aspect-ratio to reserve their box, a <script src> in the <head> carrying neither defer nor async nor type="module", @font-face without font-display, third‑party font hosts, and scripts loaded from more distinct remote domains than any reading of "minimise" defends. The hero rules are deliberately narrow — the first image inside <main>, with the header logo and the mid‑article diagram structurally excluded — so some pages get no hero finding at all, and that is returned explicitly rather than left to read as a clean result. Core Web Vitals are field data; this loads nothing and times nothing. Returns markdown plus structured output.
audit_apple_ui	Audits an iOS or macOS app against Apple's own documentation. Point it at the Xcode project directory: it reads the four surfaces a project declares configuration on — the information property list, INFOPLIST_KEY_* build settings in project.pbxproj, the entitlements plist, and each colorset's Contents.json — infers whether the target is iOS or macOS from those plus the Swift imports, and runs eight rules: a custom colorset with no luminosity: dark appearance, the deprecated UIRequiresFullScreen key (either spelling), a microphone entitlement declared under one capability and not its twin, no App Sandbox entitlement on macOS (reported as a fact about the Mac App Store channel, not as a defect), plus NavigationView, .font(.system(size:)) on iOS only, a colour written as numbers, and a Button whose whole label is one SF Symbol. Every platform‑scoped rule stays silent when the signals do not settle the question, and the report names the verdict and the signals behind it so a silence reads as the gate rather than as a result. It builds nothing, runs no simulator and takes no screenshot. Returns markdown plus structured output — findings, a severity summary, a scan block, and the longest notVisible list this server ships, every entry of it derived from a run. Directory only; a code argument, a missing path or a path that is a file comes back as an error result, not an empty audit.
audit_design_system	Is there actually a system here? Point it at real CSS/JSX and get a consistency score plus the sprawl behind it: how many distinct colors, sizes, radii, shadows and spacings are in use, which colors are indistinguishable duplicates, what's off the 4pt grid, token adoption — and a consolidation plan.
audit_generic_design	Audits for the specific defaults generated interfaces reach for — the stock Tailwind indigo/violet/purple gradient (as classes, hex, or OKLCH), Inter/Roboto/Open Sans/DM Sans/Plus Jakarta Sans as the only declared typeface on a brand surface, emoji standing in for icons, the rounded-2xl + shadow-lg + border card recipe repeated across a page, gradient‑filled heading text, an eyebrow label over every heading, the backdrop-blur + white/10 glassmorphism recipe, stock hype‑opener copy, stacked filler adverbs, and a page whose every CTA is drawn from the stock set. Every finding is a fact about the source — a class name, a phrase, a repeated structure — never a judgement about whether the result is good design; a 0–100 score counts distinct signals, not occurrences, and is itemised to rule and file:line. Pair with design_review_checklist or get_design_doc("design-critique-scoring") for the taste half. Returns markdown plus structured output — findings, a severity summary, a machine‑readable notVisible list, the itemised score, and (in directory mode) a scan block flagging a truncated read. A missing or non‑directory path comes back as an error result, not an empty audit.
audit_ux_copy	Objective copy audit — readability (Flesch), sentence length, passive voice, jargon, filler, user‑focus, weak CTAs — with flagged phrases and fixes.
list_design_knowledge	Browse the full knowledge index by category / platform.
knowledge_freshness	Reports each doc's age vs a per‑category staleness threshold, so the base can be kept current.

Resources

Beyond tools, the knowledge base is exposed as MCP resources, so clients that support them (Claude Desktop, Cursor) can @‑mention a document directly — with id autocompletion — instead of spending a tool call:

URI	What it is
saglitzdesign://index	The whole index: every document by category, with platform and last‑verified date.
saglitzdesign://doc/{id}	One knowledge document in full (96 of them). Autocompletes on id.
saglitzdesign://recipe/{component}	A component's spec plus its reference implementation in every available stack.

Visual examples

get_design_examples serves a curated library of real app/site screens. The screenshots themselves are third‑party assets and are not redistributed, so the published npm package ships the annotations and source links without the images — the tool detects this and says so rather than pretending otherwise. If you clone the repo you can rebuild the local image library; see scripts/regenerate-examples.md.

Install

Requirements: Node.js 20+.

From npm (no clone needed):

npx saglitzdesign-mcp

From source:

git clone https://github.com/HalidSaglam/saglitzdesign-mcp.git
cd saglitzdesign-mcp
npm install
npm run build

Claude Code

Register it once (via npm — no clone), available in every project:

claude mcp add --scope user saglitzdesign -- npx -y saglitzdesign-mcp

Or, if you cloned the repo, point it at the built file:

claude mcp add --scope user saglitzdesign node /absolute/path/to/saglitzdesign-mcp/dist/index.js

As a plugin

One install brings all three pieces — the MCP server, all eight skills (the seven depth skills plus the umbrella that routes into them) and a slash command for every workflow:

claude plugin marketplace add HalidSaglam/saglitzdesign-mcp
claude plugin install saglitzdesign@saglitz

The plugin declares its server as npx -y saglitzdesign-mcp@latest, so the server itself still comes from npm; the skills and the commands are files inside the plugin. claude plugin details saglitzdesign@saglitz lists what arrived.

The three ways of installing do not carry the same payload:

Install	Server, knowledge base, recipes	Skills	/saglitzdesign:… commands
claude plugin install saglitzdesign@saglitz	yes	all eight — seven depth skills plus the umbrella	one per workflow
npx saglitzdesign-mcp / claude mcp add	yes	—	—
npx skills@latest add HalidSaglam/saglitzdesign-mcp	—	all eight — seven depth skills plus the umbrella	—

The npm package's files: list covers dist/, knowledge/ and recipes/, and does not name skills/ or commands/ — so an MCP-only install has every tool and every document and none of the skill or command files. The skills CLI is the mirror image: it copies each SKILL.md into your agent and brings no server, so the tools those skills point at are not there unless you also install one. Cloning the repository gets the source of all of it — but not a runnable server: dist/ is gitignored, so a clone needs npm install && npm run build before node dist/index.js starts. Everything else in the table is a tracked file and arrives with the clone.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "saglitzdesign": {
      "command": "npx",
      "args": ["-y", "saglitzdesign-mcp"]
    }
  }
}

Cursor / other MCP clients

Run npx -y saglitzdesign-mcp over stdio (or node /absolute/path/to/dist/index.js from a clone).

Requires Node 20 or newer. It also runs on Node 18 with a warning; below that the launcher exits with an explanation instead of a syntax error.

Your client launches the server — you don't start it yourself. Run by hand it will print its status and then wait on standard input for a client that never arrives, which looks like a hang but is the server working correctly.

As skills (no MCP server)

Prefer lightweight, self-contained skills? Install the condensed SaglitzDesign skills into any skills-compatible agent:

npx skills@latest add HalidSaglam/saglitzdesign-mcp

Eight skills. saglitzdesign is the umbrella: it carries no depth guidance of its own, and routes the work into whichever of the other seven fits. Those seven — clean-interface-design, landing-page-conversion, design-review, motion-and-animation, apple-platform-design, design-system-audit, ship-quality-gate — are each standalone guidance that also points to the full MCP for depth. See skills/.

Each skill is copied into your agent and its content hash pinned in skills-lock.json, so an installed skill does not change when this repository does. Re-run the command above to pick up a new skill or an edited one.

Dev & debug

npm run dev       # run from TypeScript via tsx
npm run inspect   # open the MCP Inspector UI

Usage

Once connected, just talk to your agent naturally — it decides when to call the tools:

"Using saglitzdesign, plan the design of an iOS fitness app." → get_design_roadmap returns a 7‑phase plan with the docs to read at each step.

"Review my landing page for conversion with saglitzdesign." → design_review_checklist (landing‑page / conversion) + get_design_examples.

"How should a primary button behave on mobile?" → get_component_guidance returns specs, states, labels and anti‑patterns.

"Show me real paywall examples." → get_design_examples returns annotated screenshots.

"What's llms.txt and how do I set it up?" → seo_geo_guide (GEO) returns the tactic with a ready‑to‑use example.

Your own design rules

Point the server at a directory of your own and your documents join the base — searchable, readable, and, if you ask, part of the review checklist:

claude mcp add --scope user saglitzdesign \
  --env SAGLITZDESIGN_KNOWLEDGE_DIR=/path/to/our-design-rules \
  -- npx -y saglitzdesign-mcp

Or in claude_desktop_config.json / any MCP client, alongside command and args:

"env": { "SAGLITZDESIGN_KNOWLEDGE_DIR": "/path/to/our-design-rules" }

Several directories are allowed, separated the way PATH is on your platform.

• Your rules lead. When a search is genuinely about something you documented, your document comes first — measured by how many of the query's terms it covers, so a short house‑rules file is not buried by a long reference.
• Same id replaces ours. A file with id: buttons takes over from the built‑in one. That is deliberate, and it is announced at startup rather than happening quietly. Wherever it is served it is marked as your team's document, so nothing of yours is ever quoted as though it were sourced platform guidance.
• review: [website, saas-web-app] in the frontmatter puts the document into those project types' design_review_checklist, ahead of the curated list — the difference between your rules being findable and your rules being enforced.

Editing files inside the installed package works until npm update deletes them. Use the environment variable.

Adding to this repository

Drop a Markdown file anywhere under knowledge/ with frontmatter:

---
id: my-topic
title: "My Topic"
category: ux            # design-language | component | ux | seo | geo | pattern | craft | book | process | marketing | security
platform: both          # mobile | web | macos | both
tags: [tag1, tag2]
sources: ["https://…"]
updated: 2026-07-08
---

Content served verbatim to clients…

The server indexes every .md on startup — no rebuild needed for content changes (just restart the server). A /refresh-knowledge command (.claude/commands/) can re‑research stale docs with agents.

How the knowledge was built

Design‑language and SEO/GEO docs were researched from official documentation and current sources (cited in each file's sources). Real‑world UI patterns were studied from top apps and websites; the visual example library was curated the same way. Classic design and marketing books were distilled into original, prescriptive syntheses — no source text is reproduced.

On images: screenshot files are a local research asset. They are not included in this repository or any published package. Without them, get_design_examples gracefully degrades to descriptions plus source links. To rebuild the local image library (or add your own examples), see scripts/regenerate-examples.md.

Find it on

Also listed in the official MCP Registry and on npm.

License

MIT © 2026 Saglitz Design.

The knowledge/ documents are original syntheses with sources cited per file. Referenced screenshots are not part of this repo and may not be redistributed — see NOTICE.md.