Miko AI Descriptions & Narrate
Putting a language model where a mistake would be public
Miko AI Descriptions & Narrate is one of the apps in our own Shopify suite. Generating copy is the easy half. Making it safe to run across a thousand products, and giving the storefront a voice without a server full of audio, is the half worth writing about.
Screens, plans and the merchant guide live on the Miko product page. There is a free tier, and the paid tiers are $19, $49 and $129 a month.
Two requests that turned out to be one app
The first request was ordinary and constant: a catalogue imported from a supplier feed, every product carrying two lines of specification and no meta description, and nobody with the weeks it would take to fix by hand. The second came from a client whose customers browse with their hands full, and who asked whether a product page could simply read itself out.
They share a spine. Both need a dependable summary of a product, written in the store's language and in the store's voice. So we built one generation layer and gave it two outputs: text that goes into Shopify fields after a human approves it, and a spoken summary the storefront can play on demand.
Six things we had to solve before shipping it
Anything built on a model provider fails in ways ordinary software does not. Most of this list exists because of that.
A model answer is not trusted input
Every generation is forced through a declared response shape rather than parsed out of free text, and every field has a hand-written fallback behind it. If the model returns something unusable, the merchant gets a plain, correct sentence built from the product data instead of an error or, worse, nonsense written into a live product.
Pinned model versions rot underneath you
A specific model identifier we had pinned began answering with a not-available error in production, with nothing changed on our side, while still being listed as available. We moved both quality tiers onto the provider's rolling aliases, so the app tracks whatever is currently served at that tier instead of breaking silently the next time a version is retired.
Rules the copy is never allowed to break
Generated text may never state a price. Prices go stale the moment a merchant edits one, and under multi-currency selling the figure baked into a description is simply wrong for most of the audience. The same guardrail layer carries the house style rules and a language table, so the store's locale becomes a named language in the instruction rather than a code the model has to guess at.
Bulk work that resumes instead of restarting
A thousand-product run is processed as bounded batches, with the count of finished items acting as the cursor, so a restart mid-run never redoes committed work or double-spends a merchant's allowance. Jobs left running by a crash are reclaimed, and a job whose store uninstalled during the run fails once and terminally rather than looping forever.
A voice with no audio pipeline
The storefront reads aloud through the browser's own speech engine, so there are no audio files to generate, store, serve or pay for. The widget arrives as a theme app extension, no theme code is edited to install it, and it asks the app whether narration is actually available before it renders, so a shopper is never shown a button that would do nothing when pressed.
Fail open, or fail closed, on purpose
Each fallback direction was chosen rather than defaulted. If the availability check cannot reach us the button stays visible, because hiding a working feature over one dropped request is the worse outcome. Attribution is the opposite: removing it is a paid entitlement, so the widget confirms that with the server and keeps the credit if the answer does not arrive.
What we can add to a store you already have
If you are considering an AI feature on a real storefront, the interesting questions are the ones above. What happens when the provider answers badly, or slowly, or not at all. Who approves the output before a customer reads it. How a long job survives a deploy. How a new element reaches the storefront without a developer editing theme files every time it changes. We answered all of those in production here, on our own name, before proposing any of it to a client.
What the app does, in short
- Writes descriptions, titles, meta descriptions, alt text and tags in bulk, in the store's own language and tone
- Scans the catalogue for missing content so the gaps are visible before anything is generated
- Holds everything in review, so a person edits and approves, and any change can be reverted
- Adds a storefront Listen button through a theme app extension that plays in the browser
- Uses product data only, with no access to customer or order records
Full detail, every screen and the plans are on the Miko AI Descriptions & Narrate product page, and the merchant guide is at miko.co.nz/narrate-docs.
Two screens that show the shape of it
One faces the shopper and has to be unremarkable. One faces the merchant and has to make a large change feel small.
The storefront element. It has to look like part of the theme, install without touching theme code, and stay out of the way when it cannot play.
The bulk run. Progress is durable rather than tied to the tab, because the merchant closing the browser should never be what decides whether the work finishes.
Questions we get about this build
Tripster Developers, a Shopify agency in Auckland, New Zealand. We wrote the generation layer and its guardrails, the resumable bulk job processor, the admin interface and the storefront theme app extension, and we run the app in production ourselves.
On the product page at miko.co.nz/narrate, which carries every screen and the plans. This page is the build note that sits behind it.
Three ways. Generation is forced through a declared response shape so a field is either valid or replaced by a hand-written fallback, the prompt layer forbids stating prices because they go stale and are wrong under multi-currency, and nothing reaches a product until a person approves it, with a one-click revert afterwards.
Yes. The pattern here is the transferable part: a model call wrapped in validation and fallbacks, bulk work that resumes rather than restarts, an approval step before anything is written, and a storefront element added through a theme app extension rather than by editing theme code. Start at Shopify app development.
Yes, it is published on the Shopify App Store and maintained by us. It is one of the apps in the Miko suite, which is where we prove patterns before using them in client work.
Built by Tripster Developers
Miko AI Descriptions & Narrate is built and maintained by Tripster Developers, a certified Shopify Plus Partner working on Shopify since 2016, from Auckland, New Zealand. Narrate is the app in the Miko suite where we proved how a language model and a storefront element can be shipped safely together.
We build the same way for clients: custom Shopify themes, Shopify Plus builds, ERP and middleware integrations, and private apps for merchants who need something the App Store does not sell.
Thinking about an AI feature on your store?
Tell us what it would need to do, and we will tell you how we would make it safe.
Start a conversation