---
title: "cisegmentation"
description: "Agent skills for the Bilayers CI Segmentation workflow and its measurement databases."
canonical: https://agentpluginsdirectory.com/plugins/cisegmentation
last-updated: 2026-09-25
---

# cisegmentation
Agent skills for the Bilayers CI Segmentation workflow and its measurement databases.
- Slug: cisegmentation
- Publisher: Cellular Imaging Amsterdam UMC
- Repository: https://github.com/Cellular-Imaging-Amsterdam-UMC/cisegmentation
- Manifest: plugin.json
- Version: 0.5.0
- License: MIT
- Category (editorial): research
- Skills: 2 (analyze-cisegmentation-measurements, use-cisegmentation-workflow)
- MCP servers: 0
- Stars: 0
- Repository created: 2026-07-14
- Repository last pushed: 2026-08-10
- Publisher type: Organization
- Listing: https://agentpluginsdirectory.com/plugins/cisegmentation
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What cisegmentation does, in the publisher's words

CI Segmentation is a GPU-enabled Bilayers/BIOMERO workflow for instance segmentation from OME-Zarr to labeled OME-Zarr. It supports Cellpose 3, Cellpose-SAM v2 and the original Cellpose-SAM, PyTorch StarDist, InstanSeg, and Spotiflow with optional local mask refinement from one CUDA 12.6 environment.

- Input: one or more top-level.ome.zarr stores in /data/in, including HCS plates.
- Normal output: cisegmentation.ome.zarr in /data/out, containing native OME-Zarr label groups. By default the source store is copied with its original pixels and remains untouched; disabling Include Original Data writes a sparse, mergeable labels-only overlay and leaves the source untouched.
- Benchmark output: only benchmark_gallery_.ome.zarr.
- Axes are normalized to TCZYX; time and Z are preserved in normal runs.
- Output OME-Zarrs and schema-v5 measurement databases share an output_store_uuid; object_navigation supplies portable field, label, and ROI coordinates without embedding deployment-specific OMERO IDs.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/Cellular-Imaging-Amsterdam-UMC/cisegmentation/HEAD/README.md

## Skills

- analyze-cisegmentation-measurements: Analyze, query, and explain CI Segmentation measurement databases in DuckDB or SQLite for object morphology, per-channel intensity statistics, label sets, image or HCS plate metadata, mask relationships, focus assignments, SQL, pandas, and interpretation of CI Segmentation measurement results.
- use-cisegmentation-workflow: Configure, launch, monitor, and recover the Bilayers CI Segmentation workflow for OME-Zarr images or HCS plates using Cellpose, Cellpose-SAM, StarDist, InstanSeg, or Spotiflow.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.