Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    design md pro

    1

    Enforce design consistency and eliminate AI slop with measurable rules, HTML auditing, and Tailwind exports.

    $19

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    design md pro

    design md pro

    Example session with this skill installed

    Generate a design.md for a high-contrast dark mode dashboard. I have a primary blue #0066FF. Ask me the questions one by one.

    • Read your context and instructions
    • Compiled the design md pro

    I will now begin the interview process to generate your design.md and design.rules.json. I will ask questions one by one.

    Step 1: Brand and Logo
    What is the brand name, how should the wordmark be cased (e.g., lowercase, ALL CAPS), and are there specific backgrounds where the logo must never appear?

    Interview Progress

    • 1. Brand & Logo (Active)
    • 2. Color (Base: #0066FF)
    • 3. Typography
    • 4. Scope
    • 5. Reader & Decision
    • 6. Layout
    • 7. Text & Language
    • 8. References
    • 9. Forbidden Patterns
    • 10. Primitives
    • 11-13. Pro Rules (A11y, Responsive, Components)

    Next steps

    1. Provide the brand name and wordmark rules.
    2. Confirm if you want to use the default system font stack for typography.
    3. Prepare to define the primary "Reader" of this dashboard in Step 5.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Standard AI generation often results in generic, "slop" interfaces that ignore brand identity. This skill forces agents to follow strict, measurable design constraints by generating and enforcing a design.md and machine-readable design.rules.json.

    What it does

    • Interview & Capture conducts a 13-step structured interview to define brand, typography, layout, and component rules.
    • Automated Color Logic derives accessible surface, border, and status colors from a single accent hex using palette.py.
    • Static HTML Auditing runs audit.py against generated HTML to identify rule violations with specific error codes.
    • Framework Export generates tailwind.config.js, shadcn globals.css, or standard CSS variables directly from the rule set.
    • Slop Prevention implements a library of 52 named anti-patterns to stop agents from generating generic layouts.
    • Eval Benchmarking runs side-by-side comparisons to measure the objective impact of the design rules on output quality.

    How it works

    1. Define through a modular interview where every qualitative "vibe" is converted into a binary, observable rule.
    2. Generate the design.md documentation and design.rules.json schema to govern sub-agents.
    3. Audit existing or new HTML files using the CLI tool to find contrast, spacing, or component state errors.
    4. Export the resulting tokens into a project-ready config file for Tailwind or shadcn.

    Frameworks & tools

    Python-based CLI scripts for auditing and color math. Exports to Tailwind CSS, shadcn/ui, and standard CSS.

    Why this beats prompting it yourself

    Generic prompts like "make it modern" are subjective and fail across agent turns. This skill provides a machine-enforced boundary that catches errors in a CI-like loop, ensuring a "0 error" audit before a human ever sees the code.

    Use cases

    • Building a consistent design system for a team of autonomous frontend agents.
    • Auditing legacy HTML for accessibility and brand compliance.
    • Preventing AI-generated "slop" in landing pages or dashboards.
    • Benchmarking model performance with and without specific design constraints.

    Known limitations

    The auditor uses static analysis of HTML/CSS only. It does not perform full browser rendering or visual regression testing.

    Statik analiz yapar, render etmez. Renk türetme scripti kontrast eşiği altında hata verirse dosya üretmez. Tailwind export varsayılan ölçekleri temizler.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 14 days ago

    Requires

    Git

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed14 days ago

    What's inside

    Frequently Asked Questions

    Unlike generic prompts that agents often ignore over time, this skill creates a machine-readable `design.rules.json` and uses a Python-based auditor to verify that every line of code actually follows your brand constraints before it is finalized.

    The skill includes the 13-step Design Interview module, the `palette.py` color logic engine, the `audit.py` static analysis tool, and a library of 52 anti-pattern definitions to prevent generic 'AI slop' layouts.

    The static auditor works with any standard HTML and CSS, while the export engine provides native support for Tailwind CSS, shadcn/ui globals, and standard CSS variables.

    The auditor identifies structural, accessibility, and token violations via static analysis; however, it does not perform visual regression testing or full browser rendering, so a final human 'eyes-on' check is still recommended.

    This skill is designed for modular AI workflows and works with any LLM-based agent capable of following file-based instructions or executing CLI-based Python scripts for the auditing phase.

    New in Frontend & Web Apps