Free concept playground · no account

The Capability Avatar

Your AI skills, rendered as a character you actually earned.

The Capability Avatar

AvailableFree

Rate yourself on five AI skills. The avatar earns a tier and an archetype from your ratings — and every medallion is a receipt: it names the exact self-evaluation that unlocked it, never a cosmetic-only badge.

Promptingweight 24
75%
Evaluatingweight 20
50%
Buildingweight 24
75%
Guardingweight 16
25%
Judgingweight 16
50%
58/ 100
builderThe Unguarded Prompter
75%
50%
75%
25%
50%
Strongest: PromptingWeakest: Guarding
Every medallion is a receiptverifiable, not cosmetic

Your avatar is a Builder — “The Unguarded Prompter”. Each layer below traces back to the exact self-evaluation that earned it, so the flex is auditable, unlike a generic game badge.

  • Prompting medallion · 75%Proves: you can steer a model on purpose instead of rerolling until it looks right.
  • Building medallion · 75%Proves: you turn prompts into a system that runs without you babysitting it.
  • Evaluating medallion · 50%Proves: you measure quality against a standard, so 'good' is a number, not a feeling.
  • Judging medallion · 50%Proves: you can defend why one answer beats another — the skill behind every eval.
  • Guarding medallion · 25%Proves: you assume bad input and unsafe output, and design the guardrails first.

Next layer to unlock — Guarding: Close the safety gap: add input validation, an output schema, and a refusal path.

Share on WhatsApp
Was this playground useful?

The Capability Avatar

Rate yourself on five AI skills — prompting, evaluating, building, guarding, and judging — and the avatar turns those self-ratings into an earned capability tier (novice → practitioner → builder → architect), a 0–100 score, and a deterministic archetype drawn from your strongest and weakest axes. The moat is what it does next: instead of a cosmetic badge, every medallion is a receipt — it names the exact self-evaluation that unlocked it and tells you the one next layer worth chasing. Runs entirely in your browser (0 uploads, works offline).

How to use it

  1. Drag each of the five sliders to rate your skill 0–4, or load a preset — Fresh start, My profile, or Full architect.
  2. Watch the avatar level up live: the score, tier, and archetype update, and each medallion fills to the percentage you earned.
  3. Read the strongest and weakest callouts to see where you lead and where the gap is.
  4. Open the X-ray panel to see the receipt behind every medallion — what each earned layer actually proves — plus your single best next step.
  5. Copy or share your avatar card; a capability tier you earned is more fun to post than a generic level.

What this clears up (the fundamentals)

  • Capability is a sum of skills, not one number — prompting, evaluating, building, guarding, and judging are different muscles. A weighted score keeps a strong axis from hiding a weak one.
  • A tier should be earned, not cosmetic — the avatar never unlocks a layer for showing up. Each medallion maps to a specific self-evaluation, so the flex is auditable.
  • Your weakest axis is your next step — leveling up is targeted. Naming the lowest medallion turns "get better at AI" into one concrete rep.
  • Evaluating and judging are the quiet multipliers — most people over-index on prompting. The skill to tell a better output from a worse one, and to check it against a rubric, is what separates a builder from an architect.

Where it's used

A free, top-of-funnel gamification layer that makes progress visible — the fastest way to feel that "AI skill" is measurable rather than vibes. It's a miatz build-lab concept playable: play it here, then learn to build the weighted scoring, the eval-gated tiers, and the receipt-backed reveal yourself.

FAQ

How is the capability score calculated?

Each of the five skills is rated 0–4 and carries a weight (they sum to 100). Every skill contributes weight × (rating / 4) points; the sum, rounded, is your 0–100 score. It's a deterministic rubric — no model call, no randomness.

What are the tiers, and how do I level up?

Below 25 is novice, 25–49 is practitioner, 50–74 is builder, and 75+ is architect. You level up by raising your ratings honestly — and the reveal always points at your weakest axis as the fastest route to the next band.

What makes a medallion a "receipt"?

Every medallion links back to the exact self-evaluation that earned it — the skill, the score, and the plain-English thing it proves. That's the Demystify signature: a badge you can defend, not a cosmetic unlock.

Where does the archetype come from?

It's a short, deterministic label built from your strongest skill and your weakest one — e.g. a top prompter with the lowest judging axis reads as "The Taste-Blind Prompter." Change your ratings and the archetype changes with them.

Is anything uploaded?

No. Your ratings are scored entirely in your browser — nothing is transmitted, stored, or logged. Turn off your Wi-Fi and it still works.

Limits

A teaching self-assessment with five fixed skills and hand-set weights — it's an honest mirror of how you rate yourself, not a proctored exam or a certification. Real capability shows up in shipped work; the value here is making the axes explicit and naming the next rep. The archetypes are deliberately playful, not psychometric.

Related

Part of the Demystify Playgrounds. Explore the rest from the Playgrounds home.

Bookmark this page (Ctrl+D, or ⌘D on Mac) — it works offline the next time you need it.

Ninety playgrounds. Zero setup.

Every concept here is playable free, no account — and inside the program you learn to rebuild the machinery yourself.