What is design-with-claude CLAUDE.md?

Repository guidance for Design with Claude, a product that configures Claude Code for designers and shows command results in a browser. It documents the project, its files, and its development setup.

Use it when modifying the Design with Claude product, its MCP server, browser companion, design commands, or npm package.

It gives a coding agent the project context needed to work consistently without repeatedly inspecting the repository or guessing how its parts fit together.

How imsaif/design-with-claude configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to imsaif/design-with-claude. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/imsaif/design-with-claude/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/imsaif/design-with-claude

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/imsaif/design-with-claude/claude-md/github.svg)](https://agentmods.dev/instructions/imsaif/design-with-claude/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/imsaif/design-with-claude/claude-md"><img src="https://agentmods.dev/badge/instructions/imsaif/design-with-claude/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/imsaif/design-with-claude/claude-md"><img src="https://agentmods.dev/badge/instructions/imsaif/design-with-claude/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 3,201 This file is loaded in full into every session.

When invoked 3,201 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.03201	$0.03201
Opus 5	$0.01600	$0.01600
Sonnet 5	$0.00640	$0.00640
Haiku 4.5	$0.00320	$0.00320

Measured 9d ago against content hash af48f40e4829, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

design-with-claude CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 161 lines

How it starts

The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.

design-with-claude

Project Overview

V2 alpha live (April 2026): subscription product that configures Claude Code for designers via an MCP server, paired with a browser companion that renders command outputs live.

Live web: https://www.designwithclaude.com (start at /start)
npm package: dwic-audit (run: npx dwic-audit). Older names @imrandwc/dwic + designwithclaude are deprecated pointers to it — still owned, do not unpublish. Bare dwic unscoped is unpublishable (npm typosquat filter vs swig/twig).
Persistence: Supabase (profiles + companion_events)
See PROGRESS.md at the repo root for current state; 00-product-brief.md + 0{1,2,3,4}-*.md on Desktop for the canonical plan.

V1 (still shipping): 40+ specialized design skills as Claude Code slash commands, available as a plugin or standalone. commands/*.md stays intact — the V2 MCP server reuses those files as role prompts.

Architecture

V2 MCP server (root): TypeScript, ESM, @modelcontextprotocol/sdk over stdio. Tools live in src/tools/. Shared types in src/tools/types.ts. Six structured output kinds: palette, type-scale, spacing, component-spec, copy, markdown.
V2 install flow: npx dwic-audit setup --token=imr_xxx [--scope=user|project] writes the MCP entry. Uses claude mcp add-json when available, falls back to direct ~/.claude.json edit with timestamped backup, or .mcp.json at the cwd for project scope.
V2 gating: Pre-call /api/gating/check, post-call /api/gating/consume. Fails open when DWC_GATING is off (dev/alpha default).
V2 events: Post-call emit to /api/events. Payload shape defined in src/api-client.ts → EventPayload. The browser companion (Phase 2) subscribes via Supabase Realtime.
V1 plugin (unchanged): .claude-plugin/plugin.json manifest; commands/*.md are the knowledge base for both V1 slash commands and V2 MCP tool role prompts. /design-brief is the master command.

Key Files

package.json (root) — npm package dwic-audit, bins: dwic-audit → dist/bin/setup.js (the npx dwic-audit entry — a dispatcher: bare/flag-first runs an audit, setup/uninstall/help/version route explicitly) + dwic-mcp-server → dist/server.js (MCP server, launched by the installed .mcp.json via npx -p dwic-audit@<v> dwic-mcp-server)
src/server.ts — MCP server main; registers all tools; wraps handlers with gating + event emission
src/tools/index.ts — tool registry
src/tools/types.ts — ToolDefinition, ToolResult, output payload types
src/tools/loadPrompt.ts — reads and caches commands/*.md role prompts
src/bin/setup.ts — npx dwic-audit setup install flow
scripts/test-mcp-handshake.mjs — stdio round-trip against dist/server.js (npm run test:mcp)
scripts/test-setup-dry-run.mjs — install/uninstall against a tmp project (npm run test:setup)
.claude-plugin/plugin.json — V1 plugin manifest (unchanged)
.claude-plugin/marketplace.json — V1 marketplace catalog (unchanged)
commands/*.md — role prompts, shared by V1 slash commands and V2 MCP tools
README.md — V1 install + command reference (still current for the plugin path)
web/ — landing page; will become the V2 browser companion (Phase 2)
web/lib/dwc/{types,tokens,store}.ts — shared types + in-memory store (Symbol.for singleton survives HMR) that Phase 3 will port to Supabase
web/app/api/tokens/validate, web/app/api/gating/{check,consume}, web/app/api/events, web/app/api/events/recent — dwc API stubs consumed by the MCP server and the Phase 2 companion
web/app/api/profile/route.ts — POST mints imr_ tokens + stores onboarding answers + returns generated CLAUDE.md; GET returns stored profile
web/app/{start,profile,install,companion,upgrade}/page.tsx — V2 companion flow pages (5 screens)
web/components/companion/ — Shell, CopyButton, StartWizard, CompanionView, plus renderers/{Palette,TypeScale,Spacing,ComponentSpec,Copy,Markdown}Renderer.tsx + barrel renderers/index.tsx
web/lib/claude-md-generator.ts — extended with tone_preference for the V2 flow
web/lib/dwc/supabase.ts — lazy service_role Supabase client (server-side only; RLS blocks anon)
web/lib/dwc/store.ts — async store; dispatches to Supabase when DWC_SUPABASE_URL + DWC_SUPABASE_SERVICE_ROLE_KEY are set, in-memory Map fallback otherwise
web/supabase/{schema.sql,SETUP.md} — schema + 5-step user setup doc for Supabase project + Vercel env

Read the full file on GitHub · 161 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 161 lines · 3,201 tokens per session scan A af48f40e4829

Subscribe to this mod's changes

design-with-claude CLAUDE.md is an instructions file published in the GitHub repository imsaif/design-with-claude (11 stars, last pushed 19d ago), licensed MIT. It adds 3,201 tokens to every session, about $0.0160 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Claude Code accessibility ai-design claude-code-plugin claude-commands design-system

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

design-with-claude: Instructions file for Claude Code