Cascade
Build log · One prompt

One prompt, one turn, a working store: inside a real Cascade build

We typed one prompt into Cascade and let it run. About twelve minutes later there was a working merch store — product pages, a cart, a checkout, light and dark mode — and a full record of how it got there. Here is that record.

17 October 2026 · The Cascade team · 6 min read

Fifteen seconds of the build: the prompt goes in, the agent checks its own work, the store comes out.

The prompt

Cascade is an open-source AI app builder that runs on your machine. For this build we started from a blank project ("None — design from scratch", no template) and used GPT-6 Sol through our own API key, to show what the harness does with a strong model. The prompt opened like this:

Build Night Shift Supply — a merch store for developers who ship after midnight. It should feel like a premium independent apparel brand, not a template: confident type, generous space, real product photography, and one clear accent colour.

The rest of it — about 600 words — pinned what a designer would: a type system and palette for both themes, the page layout section by section, eight named products with prices, how the cart and checkout behave, rules for imagery, and a quality bar ("look right at 390px and 1440px, no console errors, finish by checking it in the browser"). A longer prompt is not a trick. It is the same brief you would give a person.

What the agent did

Cascade records every step of a turn, so this is not a reconstruction. In order:

  1. Planned. A planner wrote PLAN.md: the goal, the four views (home, product, checkout, confirmation), the design decisions and the data model.
  2. Loaded its playbooks for design, commerce, landing pages and app architecture.
  3. Wrote the app in batches — theme and global styles, then the product data and the logo, then the header, cards and catalog, then the product gallery, cart drawer, checkout and confirmation.
  4. Built it. The first npm run build failed; it read the error, fixed it, and built again. Later builds passed.
  5. Checked its own work in a browser. It opened its own product and checkout pages, audited them, and handed the running store to a smoke-test subagent.
  6. Closed its plan — all three tasks done.
The Cascade builder: the agent's transcript on the left, the finished store running in the live preview on the right
The builder at the end of the turn: the transcript on the left, the store running in the live preview on the right.
The run, by the numbers (from the trace)
Prompts from us1 — one agent turn, start to finish
Recorded steps (model calls and tool calls)254
Wall-clock time12 min 12 s
Tokens the model wrote34.8k
Plan tasks closed3 of 3
Activity rows: npm run build passing, Browser open /product/midnight-commit-hoodie, Browser audit, a smoke-test subagent, Browser open /checkout
The verify half of the turn: build, open its own pages, audit them, run a smoke test.
Cascade's Observatory: the span tree of the run, with the planner sub-agent, every model turn with its token counts, and every tool call
The same run in the Observatory, built into Cascade: every model turn with its tokens, every tool, and where the time went.

What it built

A store you can actually use: eight products with filters and sorting, product pages with a gallery, colour swatches and sizes (some out of stock), a cart drawer that remembers what you added and shows how far you are from free shipping, and a one-page checkout with validation and an order confirmation. It has a light and a dark theme, and it holds up on a phone.

Night Shift Supply's hero: the headline 'Goods for people who ship after midnight.' beside a photo of a developer at a desk at night
The hero, in the dark theme.
The 'Just dropped' product grid in the light theme: hoodie, tee, cap and mug, each with a price and Add to cart
The product grid in the light theme. Where no clean product photo existed, the prompt asked for drawn, on-brand mockups instead of a wrong photo.
The cart drawer: line items with quantity steppers and a free-shipping progress message
The cart drawer and its free-shipping meter.

Its own final report was candid: it said the build passed and the main flows worked in the browser, and that it had not run the 390-pixel phone check itself. We ran that afterwards — no sideways scrolling at phone width, in either theme.

What it took to get here

This was the fourth run of this prompt, and each earlier run taught us something about Cascade, not about the model:

The Code tab in Cascade showing the generated Home.tsx, formatted, with the headline text on its own line
The generated code, after turn-end formatting. The headline you see on the site is line 35.

That is the part we care about most. A build that fails is a bug report. Each of these is now a regression test, and the next person's first run gets the fixed version.

Build something with it

Cascade is free and MIT licensed. It runs on your own machine with local models through Ollama, or with your own API key.

Also on the blog: Small local models fail the same way in every harness.