# Design

> Use memi before frontend, design-system, shadcn/ui, Tailwind, Figma, UX audit, or component-registry work.

- **Type:** Skill
- **Install:** `agentstack add skill-sarveshsea-memi-design`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [sarveshsea](https://agentstack.voostack.com/s/sarveshsea)
- **Installs:** 0
- **Category:** [Content & Media](https://agentstack.voostack.com/c/content-and-media)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [sarveshsea](https://github.com/sarveshsea)
- **Source:** https://github.com/sarveshsea/memi/tree/main/plugins/memi-claude/skills/design
- **Website:** https://memoire.cv

## Install

```sh
agentstack add skill-sarveshsea-memi-design
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# memi Design Tooling

Use memi as Claude Code's local interface-understanding layer before broad UI changes. The goal is to gather product-system evidence before editing code.

## First Run

```bash
npm i -g @memi-design/cli
memi --version
memi mcp start --no-figma
```

If a repository has not been prepared for memi yet:

```bash
memi suite init --project .
memi agent install claude-code --project .
```

## UI Preflight

Run the smallest useful evidence loop before changing UI code:

```bash
memi agent brief . --agent claude-code --intent "$ARGUMENTS" --json
memi diagnose . --json
memi ux audit . --json
memi craft audit . --json
memi tokens --from ./src --report
memi shadcn doctor
```

If the work depends on a live route, replace `.` with the local URL for `diagnose`, `ux audit`, and `craft audit`.

## How To Use The Evidence

1. Read local instructions first: `CLAUDE.md`, `AGENTS.md`, README files, `.memoire/`, specs, tokens, and `memoire.agent.yaml`.
2. Treat `memi agent brief` as the preflight contract: it should shape the plan, risk list, and final verification.
3. Use `diagnose`, `ux audit`, `craft audit`, and `tokens` outputs to avoid inventing design-system facts.
4. Prefer existing components, shadcn/ui primitives, Tailwind tokens, and Atomic Design boundaries already present in the app.
5. End with the memi commands run, artifacts produced, files changed, and the next verification command.

## Registry Work

For installable UI or examples:

```bash
memi shadcn export --out public/r
memi shadcn doctor
memi publish --name @you/design-system
```

Keep registry output valid, source-controlled, and linked from the app or proof repo.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [sarveshsea](https://github.com/sarveshsea)
- **Source:** [sarveshsea/memi](https://github.com/sarveshsea/memi)
- **License:** MIT
- **Homepage:** https://memoire.cv

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-sarveshsea-memi-design
- Seller: https://agentstack.voostack.com/s/sarveshsea
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.