Skip to content
cccskills
FreePublished on GitHub

prototype

Build throwaway code to validate an uncertain approach before committing to a plan. Two branches — Logic (terminal state machine) or UI (multiple designs on one route with URL-param switcher). No persistence, no tests, single-command start. Use between analyze and plan when the approach itself is uncertain.

The English summary is being prepared. Showing the original description.

See how to install

Files included (1)

  • SKILL.md3.8 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

/supergraph:prototype

Throwaway code to validate approach. Explore fast, decide, then delete or integrate.

Announce: "🧪 /supergraph:prototype — building throwaway prototype..."

When to Use

  • Between /supergraph:analyze and /supergraph:plan when approach is uncertain
  • Multiple valid UI designs need visual comparison
  • Core logic (state machine, algorithm, data model) needs validation before full TDD
  • User says "I'm not sure which approach" or "can we try both?"

Choose Branch

Ask user (or infer from task):

Logic branch — validate state, algorithm, or data model
UI branch — validate layout, interaction, or multiple designs


Logic Branch

Goal: validate core logic in a minimal terminal program.

Rules:

  • Single file (prototype-<slug>.ts / .py / etc.)
  • No database, no HTTP, no external services — in-memory only
  • No tests — feedback is the running output
  • Single command to run: npx ts-node prototype-<slug>.ts or equivalent
  • Hard-code sample data

Structure (TypeScript example):

// prototype-<slug>.ts — THROWAWAY, do not review
type State = "idle" | "processing" | "done" | "error"
type Event = { type: "start" } | { type: "finish" } | { type: "fail" }

function transition(state: State, event: Event): State {
  // ...
}

// Run through sample cases
const cases = [ ... ]
cases.forEach(({ from, event, expected }) => {
  const result = transition(from, event)
  console.log(result === expected ? "✅" : "❌", { from, event, result, expected })
})

Run and observe:

npx ts-node prototype-<slug>.ts

UI Branch

Goal: compare multiple visual/interaction designs without committing.

Rules:

  • All designs live on ONE route: /prototype/<slug>
  • URL param ?design=A|B|C switches between designs
  • Floating control bar (fixed position) lets user switch designs visually
  • No persistence layer — mock data only
  • No auth, no real API calls
  • Single command to start: npm run dev (or project equivalent)

Structure:

app/prototype/<slug>/page.tsx   (router + floating switcher)
app/prototype/<slug>/DesignA.tsx
app/prototype/<slug>/DesignB.tsx
app/prototype/<slug>/DesignC.tsx

Floating switcher (inline):

<div style={{position:'fixed',bottom:16,right:16,zIndex:9999,background:'#fff',padding:8,borderRadius:8,boxShadow:'0 2px 8px rgba(0,0,0,.2)'}}>
  {["A","B","C"].map(d => (
    <button key={d} onClick={() => router.push(`?design=${d}`)}
      style={{margin:4, fontWeight: design===d ? 'bold' : 'normal'}}>
      {d}
    </button>
  ))}
</div>

After Prototyping

Present findings to user:

Prototype: [slug]
Branch: Logic | UI

Findings:
- [What was validated]
- [What was ruled out]
- [Surprises / unexpected behavior]

Recommendation:
- Approach: [chosen approach]
- Confidence: high | medium | low
- Reason: [evidence from prototype]

Ask: "Ready to proceed to /supergraph:plan with this approach?"

Cleanup

After approach is decided:

# Delete prototype file(s)
rm prototype-<slug>.ts
# or git stash if user wants to keep for reference

Never commit prototype code to main branch.

Rules

  • No persistence — in-memory or mock data only
  • No test suite — prototype is the test
  • Single command to run/start — no setup steps
  • Delete or integrate after decision — never leave prototype code in codebase
  • Max scope: 1 file (Logic branch) or 4 files (UI branch) — if approach is still unclear after that, escalate to user

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Risk analysis and approach selection before planning. Use when requirements are ambiguous, approaches vary, or work touches hub/bridge nodes. Skip for typo fixes.

The English summary is being prepared. Showing the original description.

datit309/supergraph21Updated Oct 18, 2026

Proactive architecture review — explore codebase structure, generate a self-contained HTML report with Mermaid diagrams and candidate improvements, then grill the findings. Use when planning a large refactor, onboarding to an unfamiliar codebase, or before a major architectural change.

The English summary is being prepared. Showing the original description.

datit309/supergraph21Updated Oct 18, 2026

Autonomous human-like browser control and professional QA/QC testing. Powered by agent-browser (Rust native CDP, zero Playwright) with ISTQB test matrix generation, automated testcase execution, console/network error tracing, and defect reporting.

The English summary is being prepared. Showing the original description.

datit309/supergraph21Updated Oct 18, 2026

Persistent token-compression mode (~75% reduction) — now always-on by default. Strips filler while keeping code exact.

The English summary is being prepared. Showing the original description.

datit309/supergraph21Updated Oct 18, 2026

Database migration best practices for schema changes, data migrations, rollbacks, and zero-downtime deployments across PostgreSQL, MySQL, and common ORMs (Prisma, Drizzle, Kysely, Django, TypeORM, golang-migrate).

The English summary is being prepared. Showing the original description.

datit309/supergraph21Updated Oct 18, 2026

Anti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-flight check.

The English summary is being prepared. Showing the original description.

datit309/supergraph21Updated Oct 18, 2026

See all skills by datit309

Report a problem with this skill