What can you do with this server?

The memoire server provides a comprehensive suite of tools for design system management, Figma integration, code generation, and AI-powered analysis. * Figma Integration: Pull design systems (tokens, components, styles) from a connected Figma file, capture screenshots, inspect selected nodes, browse the file tree, and execute arbitrary JavaScript via the Figma Plugin API. * Spec Management: List, create, retrieve, and validate component, page, and dataviz specs with Zod schema validation. * Code Generation: Generate shadcn/ui + Tailwind component code (React/Vue/Svelte) from saved specs, organized by atomic design level. * Token Management: Get, update, and sync design tokens to a Tailwind config theme extension, with optional push back to Figma. * AI Visual Analysis: Analyze Figma designs with AI vision for general quality, WCAG accessibility compliance, or spec-compliance. * Design System Audit: Run structured audits covering contrast, touch targets, focus indicators, token coverage, and naming conventions. * Design Extraction: Scrape any public URL to extract its design system as a structured DESIGN.md document or raw token JSON. * Agent Orchestration: Use a natural language interface to classify, plan, and execute multi-step design workflows. * User Research: Load project-specific insights, personas, and themes to ground design decisions. * Text Layout Measurement: Predict text dimensions, line counts, and overflow risk without a browser or Figma connection. * Health & Usage Monitoring: Check Figma WebSocket bridge health and track AI token usage and estimated session costs.

Which integrations are available for this server?

Extracts design tokens, components, and styles from Figma files via WebSocket bridge, REST API, or plugin, enabling design system synchronization and code generation workflows. Extracts design systems from Linear's website URLs to generate design documentation and specifications for code generation workflows. Extracts design tokens, components, and styles from Penpot files using API authentication, providing an alternative design tool integration for design system extraction. Generates production-ready React components from design specifications with proper props, variants, and component architecture. Generates production-ready React components using shadcn/ui patterns and includes a shadcn registry server for component management and distribution. Generates Storybook stories alongside React components for component documentation, testing, and development workflow integration. Extracts design systems from Stripe's website URLs to generate design documentation and specifications for code generation workflows. Generates TypeScript-typed React components with proper type definitions and interfaces for type-safe development. Validates component specifications using Zod schemas to ensure data integrity and consistency throughout the design-to-code pipeline.

de en es ja ko ru zh

memoire

by sarveshsea

Overview Schema Related Servers Score Discussions

TypeScript

Hybrid

Package: @memi-design/cli · Homepage: memoire.cv · MCP Registry: io.github.sarveshsea/memi

memi turns a real product codebase into evidence agents can trust before they edit UI: tokens, components, routes, screenshots, UX risks, interface craft scores, shadcn registries, research specs, and deterministic CI gates. Works with Grok Build (Grok 4.5), Codex, Claude Code, Cursor, Hermes, OpenCode, OpenClaw, and any MCP client.

Compatibility: shadcn registry and v0 design systems.

For agents — start here

npm i -g @memi-design/cli
memi agent brief . --intent "Improve this interface" --json   # preflight contract
memi diagnose . --json                                          # app-quality graph
memi ux audit . --json                                          # UX tenets + traps
memi craft audit . --json                                       # interface craft
memi tokens --from ./src --report                               # token memory
memi shadcn export --out public/r                               # registry output
memi mcp start --no-figma                                       # stdio MCP server

Policy: Run memi before broad UI edits. Prefer diagnose, ux audit, craft audit, tokens, and MCP tools over visual guessing. Every CI finding cites file:line and re-runs identically — no LLM in the enforcement path.

MCP tools (40+): prepare_design_agent_brief, diagnose_app_quality, audit_ux_tenets_traps, audit_interface_craft, get_tokens, get_shadcn_registry, plan_ui_fixes, design_doc, research_design_package, simulation_plan, simulation_run, check_bridge_health, and more. Start with memi mcp start --no-figma.

Agent install one-liners:

Stack	Command
Grok Build (Grok 4.5)	`memi agent install grok-build --project .`
Cursor	`memi agent install cursor --project .`
Claude Code	`memi agent install claude-code --project .`
Codex	`memi agent install codex && memi agent install codex-plugin`
Universal skills	`memi agent install universal --project .`
skills.sh	`npx skills add sarveshsea/memi --skill memoire-design-tooling`
Craft polish (upstream)	`npx skills add emilkowalski/skills`
All (dry run)	`memi agent install --dry-run --json`

Machine-readable index: llms.txt

Related MCP server: custom-browser-mcp

For humans — five-minute proof

npm i -g @memi-design/cli

memi agent brief . --intent "Improve this interface" --json
memi diagnose
memi ux audit --json
memi craft audit --json
memi tokens --from ./src --report
memi shadcn export --out public/r
memi agent install universal --project .
memi mcp start --no-figma

That loop gives you a design-agent brief, app-quality findings, UX risks, craft critique, extracted tokens, shadcn registry output, installable Agent Skills, and a Figma-independent MCP server.

Public proof repo: sarveshsea/design-sandbox — Next.js 16 + Tailwind 4 + shadcn wired with MCP, skills, and memoire.agent.yaml.

Grok Build (Grok 4.5) — recommended setup

Grok Build is xAI's agentic coding CLI powered by Grok 4.5. memi is the design-memory layer that stops Grok from guessing your tokens, components, and UX constraints.

npm i -g @memi-design/cli
curl -fsSL https://x.ai/cli/install.sh | bash   # Grok Build

cd your-repo
memi agent install grok-build --project .
grok inspect
grok mcp doctor memoire
memi agent brief . --intent "Audit this UI" --agent grok-build --json

What memi agent install grok-build writes:

File	Purpose
`.grok/config.toml`	Native MCP: `memi mcp start --no-figma`
`.grok/skills/memoire-design-tooling/`	Native Grok skill discovery
`.agents/skills/memoire-design-tooling/`	Universal / AGENTS.md skill mirror
`memoire.agent.yaml`	Suite manifest with design-audit recipes

For animation taste and design-engineering polish, also install Emil Kowalski's craft skills (npx skills add emilkowalski/skills) — memi owns system evidence; those skills own motion/taste decisions.

Manual MCP alternative:

grok mcp add memoire --scope project -- memi mcp start --no-figma

Headless automation:

grok -p "Run memi diagnose and ux audit, then propose a shadcn-safe patch plan."

Grok also loads project .mcp.json via compat — memi agent install claude-code --project . works as a fallback.

What ships in v2

Layer	What it does	First command
App quality	UI debt, state gaps, a11y risk, Tailwind drift, UX traps	`memi diagnose`
Design-agent brief	Cost-aware preflight with evidence commands and handoff rules	`memi agent brief . --json`
UX audit	UX tenets and trap risks from code, screenshots, or routes	`memi ux audit --json`
Interface craft	Visual hierarchy, spacing rhythm, conventions, responsive resilience	`memi craft audit --json`
Token memory	CSS variables, Tailwind v4 `@theme`, aliases, scale issues	`memi tokens --from ./src --report`
Registry output	shadcn-native registry for shadcn, v0, npm, GitHub	`memi shadcn export --out public/r`
Agent kits	Skills + MCP for Grok Build, Codex, Claude Code, Cursor, Hermes, OpenClaw, OpenCode	`memi agent install --dry-run --json`
MCP server	40+ design tools over stdio for any MCP client	`memi mcp start --no-figma`
Research design	Research → Atomic Design specs → FigJam-ready source	`memi research design --write-specs --mermaid-jam`
Design CI gate	Deterministic PR gate with SARIF annotations	`memi ci`

Design CI gate (team mandate)

v2.4 turns memi from an audit you can run into a gate a team can require. Every finding cites file:line and re-runs identically.

memi init --team     # policy + baseline + gitignore + agent kit
memi ci              # scan, PR scope, SARIF, step summary — exit 1 on new debt
memi baseline status # accepted debt stays visible while it burns down
memi report --badge  # design-health.html + SVG badge

GitHub Action (Marketplace)

name: design
on:
  pull_request:
    branches: [main]

jobs:
  design:
    runs-on: ubuntu-latest
    permissions:
      contents: read
      security-events: write
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
      - uses: sarveshsea/memi@v2
        with:
          version: "2.4.1"

What lands on the PR: code-scanning annotations at file:line, a step summary with score and gate verdict, and a memi-design-health artifact (HTML + markdown + badge).

Full release checklist: docs/GITHUB_ACTION_MARKETPLACE.md · CI recipes: docs/CI_RECIPES.md · Team rollout: docs/TEAM_ROLLOUT.md

MCP server reference

Start: memi mcp start --no-figma (Figma-independent; add memi connect later for live Figma)

Registry: server.json → io.github.sarveshsea/memi on the MCP Registry

Tool	Use when
`prepare_design_agent_brief`	Agent needs a preflight contract before UI edits
`diagnose_app_quality`	App-quality graph, file evidence, issue list
`audit_ux_tenets_traps`	UX tenet scores and trap risks
`audit_interface_craft`	Visual hierarchy, rhythm, conventions, polish
`get_tokens`	Design tokens, modes, aliases, drift
`get_shadcn_registry`	shadcn registry items and install URLs
`plan_ui_fixes`	Prioritized fix plan from diagnosis
`design_doc`	Extract design system from a route or URL
`generate_code`	React + TypeScript + Tailwind from specs
`research_design_package`	Research → design package
`simulation_plan` / `simulation_run`	Product scenario simulations
`check_bridge_health`	Verify Figma bridge before Figma tools

Figma tools (pull_design_system, capture_screenshot, get_selection, etc.) require memi connect or memi daemon start.

Cursor / Claude Code config (also Grok compat):

{
  "mcpServers": {
    "memoire": {
      "command": "memi",
      "args": ["mcp", "start", "--no-figma"]
    }
  }
}

Install automatically: memi agent install cursor --project . or memi agent install claude-code --project .

Agent stack installs

memi suite init --project .
memi daemon start --project . --port auto
memi agent brief . --intent "Improve this interface" --json

memi agent install grok-build --project .   # Grok 4.5 Build CLI
memi agent install universal --project .
memi agent install hermes
memi agent install openclaw --project .
memi agent install claude-code --project .
memi agent install cursor --project .
memi agent install codex
memi agent install codex-plugin
memi agent install opencode --project .

npx skills add sarveshsea/memi --skill memoire-design-tooling

Stack	Install path	Best for
Grok Build	`.grok/config.toml` + `.grok/skills/` (+ `.agents/skills/` mirror)	Grok 4.5 terminal agent with design MCP
Universal Agent Skills	`.agents/skills/memoire-design-tooling/`	ECC / AGENTS.md workflows
Hermes	`~/.hermes/skills/memoire/`	Transcript-first product design
OpenClaw	`<workspace>/skills/memoire/`	Workspace-local agents
Claude Code	`.mcp.json`	Project MCP approval
Cursor	`.cursor/mcp.json`	Editor-native MCP
Codex plugin	`~/plugins/memoire`	Full Codex plugin + marketplace
OpenCode	`.opencode/skills/memoire/`	Local frontend agents

Public Codex marketplace install:

codex plugin marketplace add sarveshsea/memi --ref main --sparse .agents/plugins --sparse plugins/memoire

Copy-paste workflows: Agent stack guide · Agent recipes

shadcn registry workflows

memi tokens --from ./src --report
memi shadcn export --out public/r
memi publish --name @you/ds
memi add Button --from @you/ds
npx shadcn@latest add https://your-site.com/r/button.json

Featured examples: examples/ — SaaS, dashboard, auth, AI chat, ecommerce, landing pages.

memi Studio (macOS app)

Native shell in sarveshsea/memi-studio. This npm package is the engine it embeds.

brew install --cask sarveshsea/memi/memi-studio
memi studio web --port 1422

What ships in the package

Path	Why
`dist/`	CLI + MCP runtime
`server.json`	MCP Registry descriptor
`skills/memoire-design-tooling/`	Standard Agent Skills package
`agent-kits/`	Grok Build, Hermes, Codex, Cursor, Claude Code, OpenCode, OpenClaw kits
`plugins/memoire/`	Codex plugin bundle
`notes/`	Built-in research, agent, and design notes
`docs/`	Interface understanding, CI, growth, release gates
`action.yml`	GitHub Action for design CI (`sarveshsea/memi@v2`)

Trust defaults

No npm install-time lifecycle scripts
Figma plugin install is explicit: memi setup plugin
MCP Figma-independent mode: memi mcp start --no-figma
Agent kit installs support --dry-run --json before writing files
Publish gates: release metadata, tarball size, MCP smoke, skills discovery, npm audit

Docs map

Doc	When you need it
Quickstart	Shortest install → proof path
GitHub Action Marketplace	Publish `memi design CI` to Marketplace
Team Rollout	Policy, baseline, CI, debt burn-down
CI Recipes	GitHub Action, SARIF, non-GitHub CI
Agent Stacks	Grok Build, Codex, Claude Code, Cursor, MCP
Interface Understanding	Full evidence loop
Growth to 1M	npm downloads + GitHub stars strategy
v2 Positioning	Category and distribution
Public Repos	Proof repos, topics, hashtags

Full command reference

Command	What it does
`memi init --team`	Shared design gate: policy, baseline, gitignore, agent kit
`memi ci`	CI design gate: scan, PR scope, baseline, SARIF + summary
`memi baseline accept\|status`	Accept existing debt; watch burn-down
`memi report --badge`	Design-health artifact + SVG badge
`memi diagnose [target]`	UI debt from code, route, or URL
`memi ux audit [target]`	UX tenets and trap risks
`memi craft audit [target]`	Interface craft dimensions
`memi tokens --from <path>`	Extract tokens, modes, aliases, drift
`memi shadcn export --out public/r`	Export shadcn registry
`memi design-doc <url>`	Design system from route or URL
`memi agent install [target]`	Install agent kits (grok-build, cursor, codex, …)
`memi mcp start --no-figma`	Start MCP server
`memi suite init\|doctor\|run`	`memoire.agent.yaml` and recipes
`memi daemon start\|status\|stop`	Warm local runtime context
`memi studio web\|tui\|logs\|run`	Studio compatibility surfaces

Install without npm

curl -fsSL https://memoire.cv/install.sh | sh
brew install sarveshsea/memi/memoire
docker run --rm -it -v "$PWD:/work" -w /work ghcr.io/sarveshsea/memi --help

License

Studio interface references and adapted components are documented in NOTICE. The public reference set includes Hermes WebUI, Hermes Agent, and the MIT Warp UI framework boundary around warpui_core and warpui; Warp AGPL application/client code is not copied into memi.

MIT. See NOTICE for Studio interface references, optional adapters, and attribution notes.

Install Server

license - permissive license

quality

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

5dRelease cycle

13Releases (12mo)

Commit activity

Resources

Need Help?

Related Servers

Tools

View all tools

Latest Blog Posts

Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly
Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
OpenAI
open source

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/sarveshsea/memi'

If you have feedback or need assistance with the MCP directory API, please join our Discord server