IntrinsentGet in touch

ChatGPT apps for discovery-driven products.

Intrinsent designs and builds ChatGPT apps that connect your marketplace, catalog, directory, or data product directly into the conversation.

Show me some important paintings from the last few centuries with water as the main focus.

I found important works where water is a major subject — open any one for details and sources.

Art ExplorerArt Explorer

Paintings to explore

Sourced from open collections · sorted by notability

6 results
The Great Wave off Kanagawa

The Great Wave off Kanagawa

Katsushika Hokusai · c. 1831

Open
Water Lilies

Water Lilies

Claude Monet · 1906

Open
Impression, Sunrise

Impression, Sunrise

Claude Monet · 1872

Open
The Fighting Temeraire

The Fighting Temeraire

J. M. W. Turner · 1839

Open
The Gulf Stream

The Gulf Stream

Winslow Homer · 1899

Open
Storm over the Black Sea

Storm over the Black Sea

Ivan Aivazovsky · 1893

Open

Artworks via Wikimedia Commons (public domain).

Ask anything

A ChatGPT app we built — view the live example →

01 / Ask

Search is moving into the conversation.

People are starting discovery with natural language, not a filter panel. The first job is choosing the search moment from your product that belongs in ChatGPT.

|
A natural-language request starts the flow. No filters, no search form, no new interface to learn.
Ask anything
02 / Structure

Turn intent into a product-shaped request.

ChatGPT extracts the useful parts and calls your MCP server with typed inputs your product can actually answer.

Prompt decomposition

Show me some important paintings from the last few centuries with water as the main focus.
paintings→ typesome→ limitimportant→ sortBylast few centuries→ yearMinwater as the main focus→ subjects
becomes typed arguments

MCP tool call

searchPaintings
{
"name": "searchPaintings",
"arguments": {
"type": "painting",
"limit": 6,
"sortBy": "notability",
"yearMin": 1700,
"subjects": ["water", "sea", "wave", "river", "lake", "pond"]
}
}
03 / Render

Show the answer like a product, not a paragraph.

Results should feel native in ChatGPT: cards, carousels, detail views, empty states, source links, and the right amount of context for the next decision.

Show me some important paintings from the last few centuries with water as the main focus.

I found six notable works where water is a major visual subject. You can browse them directly here.

Results in chat

Image-first cards · detail views · actions

6 results
The Great Wave off Kanagawa

The Great Wave off Kanagawa

Katsushika Hokusai · c. 1831

Open
Ask anything
04 / Convert

Send qualified intent back to your product.

The ChatGPT surface is not the destination. It opens a new channel, qualifies the request, and hands users back to your product when they are ready to act.

Open the Monet result.

Here is the selected result with a direct handoff to the source page.

Impression, Sunrise by Claude Monet

Selected result

Impression, Sunrise

Claude Monet · 1872

Discovery happens in ChatGPT. The handoff takes the user to the product page, source page, checkout, booking flow, or whatever action comes next.

wikipedia.org/wiki/Impression,_Sunrise
Open on Wikipedia
Ask anything

We build the product layer between ChatGPT and your data.

We turn natural-language intent into typed MCP calls, custom UI inside ChatGPT, and a clear path back to your product when the user is ready.

MCP serversStructured tool schemasInline ChatGPT UIProduct handoffs

Want to work with us?

Get in touch