HN Showcase
Back to browse
GitHub Repository

An AI coding skill that transforms project planning into visual, interactive HTML decision documents

53 stars

A Claude Code skill that renders decisions as interactive HTML pages

by jontomato·Mar 5, 2026·5 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidBig BrainShip It

Visual decision trees beat text prompts for UI/UX choices, but needs multi-agent AI tooling to shine.

Strengths
  • •Four-option visual comparisons with rendered CSS mockups solve the 'navbar vs sidebar' problem that text descriptions can't handle
  • •Decision history persists as HTML files—you can revisit and change past choices, updating downstream plans automatically
  • •Self-contained HTML output means zero infrastructure; decisions live in `.decisions/` folder alongside your code
Weaknesses
  • •Tied to Claude Code ecosystem; won't work with other LLM planning tools or generic MCP servers
  • •Early-stage rough edges acknowledged by author; no evidence of production usage or iteration feedback
Target Audience

Full-stack developers, product managers, teams using Claude Code for project planning

Similar To

Continue.dev decision logging · GitHub Copilot workspace planning · Cursor project templates

Post Description

When AI coding tools help you plan a project, they describe your options in text and ask you to pick. That works fine for technical choices but falls apart for anything visual. "A sticky navbar with a hamburger menu" vs "a sidebar with collapsible sections" is hard to evaluate without seeing them.

I built a Claude Code skill that generates a self-contained HTML page for each decision point and opens it in the browser. Each page has four options with visual previews (rendered CSS mockups for UI decisions, flow diagrams for interactions, architecture diagrams for technical choices), a comparison table, and a recommendation. You pick one, it records the choice, and moves on. At the end you get a standard implementation plan.

All states live in a .decisions/ folder as HTML files and a JSON manifest. There's a landing page that shows every decision and its status. You can change past decisions and it updates everything.

Tradeoffs worth knowing about: it's meaningfully slower than text-based planning. Each decision is a full HTML file generation. It uses more tokens. The visual previews are CSS approximations, not pixel-perfect mockups. For small projects or projects where you already know what you want, it's overkill.

The source is one file (SKILL.md) that acts as a prompt template. No dependencies, no build step, no runtime beyond the AI itself.

Anyway, give it a try. Hope you all like it.

Similar Projects