Connect your tool

The twelve scans Scroll Lens reads are open data, and each one reads straight from its URL in your own tool. The page itself is also an API and an MCP server.

A scan it reads, example PHerc0139 2.399 um

CC BY-NC 4.0 (Vesuvius Challenge open data) · The other eleven are listed by list_scans; all are read live, never copied here. Scroll Lens reads level 0. · browse

https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0139/volumes/20260102150214-2.399um-0.2m-78keV-masked.zarr

Python (zarr)tested 2026-10-24

# pip install "zarr>=3" fsspec aiohttp
import zarr

root = zarr.open_group("https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0139/volumes/20260102150214-2.399um-0.2m-78keV-masked.zarr", mode="r")  # OME-Zarr, zarr v2
a = root["3"]  # level 0 is full resolution; level L is 2**L times coarser
print(a.shape, a.dtype, a.chunks)
block = a[4810:4874, 1657:1721, 1657:1721]  # (z, y, x); reads only the chunks it touches

villa's vesuvius packagetested 2026-10-24

# pip install "vesuvius[volume-only]"
from vesuvius import Volume

v = Volume(type="zarr", path="https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0139/volumes/20260102150214-2.399um-0.2m-78keV-masked.zarr/3")  # one pyramid level
block = v[4810:4874, 1657:1721, 1657:1721]

VC3Dfrom villa's source

In VC3D (villa's volume-cartographer), open or create a project, then
File > Attach Remote Zarr... and paste:

https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0139/volumes/20260102150214-2.399um-0.2m-78keV-masked.zarr

It accepts http, https and s3 OME-Zarr roots.

neuroglancertested 2026-10-24

Open in neuroglancer

https://neuroglancer-demo.appspot.com/#!%7B%22layers%22%3A%5B%7B%22type%22%3A%22image%22%2C%22source%22%3A%22zarr%3A%2F%2Fhttps%3A%2F%2Fvesuvius-challenge-open-data.s3.us-east-1.amazonaws.com%2FPHerc0139%2Fvolumes%2F20260102150214-2.399um-0.2m-78keV-masked.zarr%2F%22%2C%22name%22%3A%22A%20scan%20it%20reads%2C%20example%20PHerc0139%202.399%20um%22%7D%5D%2C%22layout%22%3A%224panel%22%7D

curltested 2026-10-24

curl -s https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0139/volumes/20260102150214-2.399um-0.2m-78keV-masked.zarr/.zattrs            # OME multiscales metadata
curl -s https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0139/volumes/20260102150214-2.399um-0.2m-78keV-masked.zarr/3/.zarray     # one level: shape, chunks, dtype, codec

Call Scroll Lens from code or an agent

Code Apache 2.0; models as listed by list_scans; scans CC BY-NC 4.0 · A new point uses the caller's ZeroGPU minutes. Model output, not a reading.

https://huggingface.co/spaces/SharpMeow/scroll-lens

Python /list_gallerytested 2026-10-24

# pip install gradio_client
from gradio_client import Client

client = Client("SharpMeow/scroll-lens")
result = client.predict(api_name="/list_gallery")

Python /list_scanstested 2026-10-24

# pip install gradio_client
from gradio_client import Client

client = Client("SharpMeow/scroll-lens")
result = client.predict(api_name="/list_scans")

Python /random_pointtested 2026-10-24

# pip install gradio_client
from gradio_client import Client

client = Client("SharpMeow/scroll-lens")
result = client.predict(scan="PHerc0139", seed=0, api_name="/random_point")

curl /list_gallerytested 2026-10-24

URL=https://sharpmeow-scroll-lens.hf.space
EVENT=$(curl -s -X POST $URL/gradio_api/call/list_gallery -H 'Content-Type: application/json' \
  -d '{"data": []}' | python3 -c 'import json,sys; print(json.load(sys.stdin)["event_id"])')
curl -sN $URL/gradio_api/call/list_gallery/$EVENT

MCP (Claude, Cursor and other agents)tested 2026-10-24

{
  "mcpServers": {
    "scroll-lens": {
      "url": "https://sharpmeow-scroll-lens.hf.space/gradio_api/mcp/sse"
    }
  }
}

Tool list: https://sharpmeow-scroll-lens.hf.space/gradio_api/mcp/schema