Rumpelstiltskin Dance AI: Locking a Tip-Toe Barn Meme Template

Posted on Oct 31 • Originally published at rumpelstiltskindance.video

How I shipped a 1–2 photo → tip-toe barn meme dance on Cloudflare Workers — template lock, guest Stripe, Seedance via TryAllAPI.

rumpelstiltskindance.video turns 1–2 adult photos into a tip-toe / “Wing In My Jawwdinz”-adjacent meme dance short with sound. Independent fan tool (2026), 18+. The homepage does not ask users to storyboard; it locks the motion template and sells the render.

Angle for this post: template lock done with an optional second dancer — similar family to party sites, different farm/barn beat. Prices and model ids change; treat live site / console as source of truth.

Copy on-site refuses celebrity / MV stills. That is load-bearing product policy for face tools — not a legal afterthought.

Live pricing verified: 500 credits · $8.88 · 720p, with multi-packs (e.g. $16.99, larger tiers). Guest Stripe. Always re-check the create button.

Meme dances are recognizable moves. If you let users rewrite the choreography prompt, you get mush that does not match the sample. Locking the motion:

lead (+ optional friend) → tip-toe template → check credits (else Stripe) → Worker creates Seedance task via TryAllAPI → poll until MP4 URL → store / show → on failure: refund credits Enter fullscreen mode Exit fullscreen mode Why a video API aggregator Operational checklist after register / top-up:

Conceptually the call shape looks like this (field names follow the aggregator’s v3 contents API; always re-check current docs):

POST /api/v3/contents/generations/tasks Authorization: Bearer <TRYALLAPI_API_KEY> Content-Type: application/json Accept-Encoding: identity GET /api/v3/contents/generations/tasks/{id} Enter fullscreen mode Exit fullscreen mode I send Accept-Encoding: identity from the Worker because some gateways behave badly with transparent decompression. Model ids used in production look like Seedance 2.x family strings — treat console ids as source of truth, not this paragraph.

A few details that bit me across this family of sites:

Rough mental model for upstream cost (illustrative only — use your own usage fields and invoices):

tokens ≈ duration_sec × width × height × fps / 1024 cost ≈ tokens × price_per_million_tokens Enter fullscreen mode Exit fullscreen mode Price the site after you measure a few real runs with retry margin, Stripe fees, and failed-job refunds — not from a spreadsheet fantasy of 0% failure.

Map lead vs friend roles consistently. Rehost short-lived result URLs. Separate keys if you operate a matrix of dance domains.

Demo UIs are easy. The line between toy and product is:

Two books of account stay separate: USD balance at TryAllAPI (your cost) vs site credits (what you sold). Mixing them in one mental ledger is how you underprice or invent “mysterious” losses.

When several products share one Stripe account, isolate webhooks by endpoint or by metadata ( site / product ). Callbacks pointed at the wrong apex domain are a classic multi-site footgun.

Secrets stay in Workers: TRYALLAPI_API_KEY, STRIPE_SECRET_KEY, webhook secret, auth secrets. The browser only ever sees publishable Stripe keys if needed.

pnpm install # wire D1 database_id, production env, custom domain in wrangler pnpm cf:deploy Enter fullscreen mode Exit fullscreen mode Before calling a deploy “done”:

Production origin: https://rumpelstiltskindance.video.

Face-based generation is not “just another SaaS CRUD app”:

Those constraints belong in the product, not in a footnote you add after the first dispute email.

Consumer: watch the sample on rumpelstiltskindance.video, upload authorized photos, pay once. Builders: TryAllAPI, prove create+poll, then wrap credits. Lock the tip-toe; meter the GPU.