Dg Discover logo

Dg Discover

Community
FReptar0
dg-discover

Research a business and generate a tailored DESIGN.md. Use when the user wants to redesign a website, create a design system, or build a web presence for a specific business. Investigates the business model, analyzes competitors, understands the audience, and only generates when confident enough. The recommended entry point for any real business project.

Overview

PublisherFReptar0
Repositorydesign-guard
Skill namedg-discover
Stars
12
Forks
4
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by FReptar0 on GitHub. Read the source before you install it.

Installation

Install the Dg Discover AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/FReptar0/design-guard/tree/main/.claude/skills/dg-discover
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/FReptar0/design-guard.git /tmp/design-guard
mkdir -p .claude/skills
cp -r /tmp/design-guard/.claude/skills/dg-discover .claude/skills/dg-discover
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Dg Discover in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Dg Discover on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Dg Discover is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Research a business and generate a tailored DESIGN.md with autonomous investigation.

Core Principle: Understand Before Designing

NEVER generate a DESIGN.md until you understand the business. A beautiful design for the wrong business model is worse than no design at all.

Phase 1: Gather Context (REQUIRED — cannot skip)

Before ANY research or generation, establish these fundamentals:

  1. What the business IS:

    • If only a company name is given, ask: "What does [company] do? What products or services do they offer?"
    • If unclear whether physical or digital: "Do they sell online, in physical stores, or both?"
  2. Who the customers are:

    • "Who are their primary customers? (families, developers, enterprises, etc.)"
  3. What the website should achieve:

    • "What is the main goal of this website? (drive store visits, sell online, generate leads, inform)"
    • This is CRITICAL — it determines the entire page structure
  4. Current web presence:

    • "Do they have a current website? What URL?"

If the user provides a detailed brief, extract answers from it. But if any critical info is missing (especially #1 and #3), ASK before proceeding. Do NOT assume.

Context Isolation

You are researching a specific target business, NOT the host project (Design Guard).

  • CLAUDE.md describes how Design Guard works as a development tool. It does NOT describe the target business. Do not confuse the two.
  • All research findings, business model analysis, and audience insights must be about the target business, not about Design Guard.
  • When saving to .dg-research/latest.json, the brief.companyName field MUST match the target business name, never "Design Guard" (unless the user explicitly asked to research Design Guard itself).
  • If you are unsure whether a fact comes from the target business or from CLAUDE.md project context, discard it and research again using WebSearch.
  • Do NOT describe the target business as "open source", "CLI tool", "design intelligence", or any other phrase from CLAUDE.md's Project Overview.

Phase 2: Research (autonomous)

Once you have the basics, research to validate and enrich:

  1. Search for the business: Use WebSearch for "[company name]" and "[company name] business model"

    • Determine: Is this physical retail? E-commerce? SaaS? Service?
    • Find: Store count, locations, revenue model, market position
    • Find: Slogan, tagline, brand messaging
  2. Analyze current website (if URL available): Use WebFetch on the homepage

    • Look at navigation items — what do they tell you about the business?
      • "Sucursales/Stores/Locations" → physical retail
      • "Cart/Checkout/Shop" → e-commerce
      • "Login/Dashboard/API" → SaaS
      • "Contact/Book/Schedule" → service
    • Extract current brand colors from CSS
    • Identify current fonts
    • Note CTAs — what action does the site want visitors to take?
  3. Research competitors: Use WebSearch for "[company name] competitors" or "[industry] [market] companies"

    • Analyze 1-2 competitor websites via WebFetch
    • Note their color schemes, fonts, and layout patterns
    • Identify what to differentiate FROM

Confidence Check (MUST pass before generating)

Rate your confidence on each dimension (0-100):

DimensionMinimum RequiredWhat It Means
Business model>= 80You know what they do and how they make money
Website purpose>= 80You know what the site should achieve
Audience>= 60You know who the customers are
Visual identity>= 40You have colors/fonts (can use presets if needed)
Competitive>= 30You know at least who competitors are

Weighted average must be >= 70 to proceed.

If below threshold:

  • Ask the user for missing information
  • Do more research
  • DO NOT proceed until confident

Report your confidence: "My understanding confidence: [X]%. Business model: [type]. Proceeding to generate."

Phase 3: Generate DESIGN.md

  1. Generate DESIGN.md with 9 effective sections (Section 1 expanded):

Section 1: Visual Theme & Business Context

This section is the most important — Stitch reads it on every generation.

Include:

  • 2-3 sentences describing the visual direction
  • Business Model: What the company IS (e.g., "Physical retail grocery chain with 3,300+ stores")
  • Website Purpose: What the site should DO (e.g., "Drive foot traffic to stores. NOT an e-commerce site.")
  • Primary User Goals: Numbered list of what visitors want to accomplish
  • Key Page Elements: What the site MUST have (e.g., "Store locator, weekly deals, category browse")
  • Avoid: What the site must NOT have (e.g., "Shopping cart, checkout, add-to-cart buttons")

Sections 2-7: Standard design tokens

Colors, typography, spacing, components, iconography, imagery — as before.

Section 8: Do's and Don'ts

Standard anti-slop rules PLUS business-model-specific rules:

  • Physical retail: "Do make store locator the primary CTA", "Don't add shopping cart"
  • SaaS: "Do show product screenshots", "Don't use generic feature icons"
  • Service: "Do make booking/contact primary", "Don't hide contact info"

Phase 4: Validate & Present

  1. Before writing, self-check:

    • Does Section 1 include business model context? (REQUIRED)
    • Does it explicitly state what the site is NOT? (REQUIRED for physical retail)
    • Are Do's/Don'ts business-specific, not just visual? (REQUIRED)
    • Are all colors hex values? (REQUIRED)
    • Is it under 3000 tokens? (REQUIRED)
    • Does it reference the actual audience? (REQUIRED)
  2. Present findings to user:

    • "Business type: [type]"
    • "Website purpose: [purpose]"
    • "Brand colors found: [colors]"
    • "Competitor differentiation: [approach]"
    • "Confidence: [X]%"
  3. Write DESIGN.md and save research to .dg-research/.

  4. Suggest first screen prompt that ALIGNS with the business model:

    • Physical retail: "A landing page with hero, store locator as primary CTA, weekly deals section, and category browse"
    • SaaS: "A landing page with product demo hero, feature section, pricing tiers, and signup CTA"
    • Service: "A landing page with service overview, booking CTA, testimonials, and team section"

Guardrails

  • NEVER generate before understanding the business model
  • NEVER assume e-commerce — most businesses in the world are NOT online stores
  • ALWAYS include business context in Section 1 of DESIGN.md
  • ALWAYS include what the site should NOT have (notFeatures)
  • If unsure about the business model, ASK the user
  • The DESIGN.md is not just a style guide — it's the business context that prevents wrong designs

Frequently asked questions

What does the Dg Discover AI skill do?

Research a business and generate a tailored DESIGN.md. Use when the user wants to redesign a website, create a design system, or build a web presence for a specific business. Investigates the business model, analyzes competitors, understands the audience, and only generates when confident enough. The recommended entry point for any real business project.

Why use Dg Discover on TypingMind?

Because you install it once and use it with any model. Dg Discover is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Dg Discover in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/FReptar0/design-guard/tree/main/.claude/skills/dg-discover. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Dg Discover?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Dg Discover?

As many as you like. As long as a model supports skills, you can use Dg Discover with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Dg Discover AI skill free?

Yes. It is published on GitHub by FReptar0 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇