CONNECT.md

Connect your tool: Ink Lens

The sample segments Ink Lens reads are open data, and each one reads straight from its URL in your own tool. The page is also an API and an MCP server.

Checked 2026-10-24. Generated from connect.json by kit spaceconnect in ChaseHendrick/Scrolls.

Sample layers, PHerc0841 w00 (9.366 um surface volume)

https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/surface-volumes/9.366um-1.2m-113keV-volume-20250821151531.zarr

License: CC BY-NC 4.0 (Vesuvius Challenge open data).

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/PHerc0841/segments/20260220213127-w00/surface-volumes/9.366um-1.2m-113keV-volume-20250821151531.zarr", mode="r")  # OME-Zarr, zarr v2
a = root["0"]  # level 0 is full resolution; level L is 2**L times coarser
print(a.shape, a.dtype, a.chunks)
block = a[0:64, 2800:2864, 2900:2964]  # (z, y, x); reads only the chunks it touches

villa's vesuvius package (tested 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/PHerc0841/segments/20260220213127-w00/surface-volumes/9.366um-1.2m-113keV-volume-20250821151531.zarr/0")  # one pyramid level
block = v[0:64, 2800:2864, 2900:2964]

curl (tested 2026-10-24)

curl -s https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/surface-volumes/9.366um-1.2m-113keV-volume-20250821151531.zarr/.zattrs            # OME multiscales metadata
curl -s https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/surface-volumes/9.366um-1.2m-113keV-volume-20250821151531.zarr/0/.zarray     # one level: shape, chunks, dtype, codec

Its surface (tifxyz)

https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/mesh/20260220213127-on-20250821151531-9.366um.tifxyz

License: CC BY-NC 4.0 (Vesuvius Challenge open data).

Python (numpy, tifffile) (tested 2026-10-24)

# pip install numpy tifffile
import io, json, urllib.request
import numpy as np, tifffile

T = "https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/mesh/20260220213127-on-20250821151531-9.366um.tifxyz"
meta = json.load(urllib.request.urlopen(T + "/meta.json"))  # scale, bbox, format
get = lambda c: tifffile.imread(io.BytesIO(urllib.request.urlopen(f"{T}/{c}.tif").read()))
xyz = np.stack([get(c) for c in "xyz"], -1)  # xyz[i, j] = (x, y, z) in scan voxels; -1 = no surface

The same segment's 2.4 um surface volume (what every model reads in Compare models)

https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/surface-volumes/2.403um-0.22m-77keV-volume-20260319124803.zarr

License: CC BY-NC 4.0 (Vesuvius Challenge open data).

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/PHerc0841/segments/20260220213127-w00/surface-volumes/2.403um-0.22m-77keV-volume-20260319124803.zarr", mode="r")  # OME-Zarr, zarr v2
a = root["0"]  # level 0 is full resolution; level L is 2**L times coarser
print(a.shape, a.dtype, a.chunks)
block = a[0:64, 11200:11264, 11600:11664]  # (z, y, x); reads only the chunks it touches

villa's vesuvius package (tested 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/PHerc0841/segments/20260220213127-w00/surface-volumes/2.403um-0.22m-77keV-volume-20260319124803.zarr/0")  # one pyramid level
block = v[0:64, 11200:11264, 11600:11664]

curl (tested 2026-10-24)

curl -s https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/surface-volumes/2.403um-0.22m-77keV-volume-20260319124803.zarr/.zattrs            # OME multiscales metadata
curl -s https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/surface-volumes/2.403um-0.22m-77keV-volume-20260319124803.zarr/0/.zarray     # one level: shape, chunks, dtype, codec

Its 2.4 um surface (tifxyz), the mesh the 3D models are read onto

https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/mesh/20260220213127-on-20260319124803-2.403um.tifxyz

License: CC BY-NC 4.0 (Vesuvius Challenge open data).

Python (numpy, tifffile) (tested 2026-10-24)

# pip install numpy tifffile
import io, json, urllib.request
import numpy as np, tifffile

T = "https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/PHerc0841/segments/20260220213127-w00/mesh/20260220213127-on-20260319124803-2.403um.tifxyz"
meta = json.load(urllib.request.urlopen(T + "/meta.json"))  # scale, bbox, format
get = lambda c: tifffile.imread(io.BytesIO(urllib.request.urlopen(f"{T}/{c}.tif").read()))
xyz = np.stack([get(c) for c in "xyz"], -1)  # xyz[i, j] = (x, y, z) in scan voxels; -1 = no surface

Call Ink Lens from code or an agent

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

License: Code Apache 2.0; models MIT (each credited in list_models); scans CC BY-NC 4.0. Checked live on 24 Oct 2026: /list_samples and /lens_sample answered (gradio_client, the 0841-w00 sample). Model output, not a reading.

Python /lens_sample (tested 2026-10-24)

# pip install gradio_client
from gradio_client import Client

client = Client("SharpMeow/ink-lens")
result = client.predict(sample_id="0841-w00", size=640, reader="d9v2", windows=3, api_name="/lens_sample")

Python /list_samples (tested 2026-10-24)

# pip install gradio_client
from gradio_client import Client

client = Client("SharpMeow/ink-lens")
result = client.predict(api_name="/list_samples")

Python /compare_sample (tested 2026-10-24)

# pip install gradio_client
from gradio_client import Client

client = Client("SharpMeow/ink-lens")
result = client.predict(sample_id="0841-w00", models="ink_9um-s42,reader-v3,hecate-9.6", size=384, scan="2.4", centers="0", steps="1", api_name="/compare_sample")

Python /list_models (tested 2026-10-24)

# pip install gradio_client
from gradio_client import Client

client = Client("SharpMeow/ink-lens")
result = client.predict(api_name="/list_models")

curl /lens_sample (tested 2026-10-24)

URL=https://sharpmeow-ink-lens.hf.space
EVENT=$(curl -s -X POST $URL/gradio_api/call/lens_sample -H 'Content-Type: application/json' \
  -d '{"data": ["0841-w00", 640, "d9v2", 3]}' | python3 -c 'import json,sys; print(json.load(sys.stdin)["event_id"])')
curl -sN $URL/gradio_api/call/lens_sample/$EVENT

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

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

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