Cat Hero Studio

Live

Count every cat in your photo and turn each one into its own superhero.

musemuse-cat-hero-studio.vercel.appOpen

Get it

Connect it to any app that speaks MCP. It has 2 tools: segment_cats, create_superhero_cat. No sign-in.

Muse app: add this URL as a custom connector.
Claude Code: claude mcp add --transport http cat-hero-studio https://muse-muse.vercel.app/api/mcp/cat-hero-studio
Anything else: add it as a remote server over streamable HTTP.

Each line is what the check ran on, and no more.

This is a model's opinion, not proof.

Nothing here should be taken as tested.

Not checked Whether its answers are true
Not checked ChatGPT: only a person with an account can test it
Not checked Installing the Claude plugin or the Muse gadget pack on a real device
Not checked No Prove it run: nothing it comes with was run for real

A text note asking for a site that uses segmentation to count cats in a photo and makes a separate superhero image for each cat.

SAM 3.1Finds and outlines every cat in the uploaded photo from the prompt cat, so the product can count and crop each cat.
Muse ImageTurns each cropped cat into a superhero portrait with a cape, mask and heroic pose while keeping the cat recognizable.

Counts come from one photo and are estimates. Cats that are hidden, blurry, or partly behind furniture may be missed or merged.
Segmentation finds at most 16 cats in one photo.
Superhero portraits are generated interpretations. Fur pattern, face, and pose may change a little from the original cat.
Very crowded or overlapping cats may produce less accurate crops and portraits.
Results are not saved on a server. Refreshing the page may clear them.

Add it in the Muse app as a custom connector. It exposes 2 actions: segment_cats, create_superhero_cat.

Muse Code made 62 tool calls and 23 edits across 23 files (+679 −49).
npm run verify passed on the first try.
Screenshot judge: pass. Final screen shows the count of 3 cats with the original photo outlined and numbered 1 to 3.
Built in 9 min for about $3.32 in model calls (est.).
Publishing is off for this run (it's a local run, and a published product couldn't reach this machine's gateway), so the product didn't get its own link. The preview, checks and demo above ran for real.
How it was built: 62 tool calls, 23 edits touching 23 files { <read-skill-result name="meta-ui" status="ok"> <read-skill-result name="meta-sam" status="ok"> <read-skill-result name="meta-image" status="ok"> <read-skill-result name="product-ui" status="ok"> <read-skill-result name="verify" status="ok"> { <read-skill-result name="bundled:taste" status="ok"> { {"items":4,"ok":true,"revision":1} { Edited `lib/server/image.ts` <read-skill-result name="bundled:durable-test-collateral" status="ok"> Edited `app/api/actions/segment_cats/route.ts` Edited `app/api/actions/create_superhero_cat/route.ts` Edited `app/api/sample-photo/route.ts` { Edited `app/api/actions/segment_cats/route.ts` {"items":4,"ok":true,"revision":2}
Imported from a local run. Publishing it to its own link now.