LABSCO SUMMARY

Two run modes, and the choice is repeatability against reach.

Tool mode runs a fixed set of scanpy functions, so the same request behaves the same way every time; code mode hands the model full Python and covers anything the predefined functions don't, at the cost of varying between runs. If the analysis ends up in a paper, the authors ask that you cite scanpy, LIANA+, decoupleR and CellRank 2.

What it is

A Python server that puts a scRNA-seq workflow behind natural language: reading and writing data, preprocessing, clustering, differential expression, plotting, cell-cell communication, pseudotime and enrichment.

What you get

An IO module for reading and writing scRNA-seq data
Preprocessing: filtering, quality control, normalization, scaling, highly-variable genes, PCA and neighbors
Analysis: clustering and differential expression, plus cell-cell communication, pseudotime and enrichment
Plots: violin plots, heatmaps and dotplots
Two run modes — tool mode calls predefined functions; code mode gives the model a Jupyter backend and returns a complete notebook with executable code, results and figures

Requirements

pip install scmcp, then scmcp run --run-mode tool or scmcp run --run-mode code. For remote use, scmcp run --transport shttp --port 8000 and point the client at http://localhost:8000/mcp.

Setup effort

One command — pip install scmcp

See installation options

USE CASES

01

Getting from a raw matrix to clusters

Filter, normalize, pick highly-variable genes, run PCA and neighbors, then cluster — described in sentences rather than assembled by hand.

02

Asking which genes separate two clusters

Run differential expression on the clustering you just produced and get the result back in the same conversation.

03

An analysis the predefined functions don't cover

Switch to code mode; the Jupyter backend executes generated Python and any library you need.

04

Handing a collaborator something reproducible

Code mode emits a full notebook — the code that ran, the results and the figures — instead of a chat transcript.

BEFORE YOU INSTALL

What to know first

Requires

pip on PATH

INSTALLATION

Connect your client

Add the block to your client, or run the command directly.

Server setup

Configuration file

JSON

{
  "mcpServers": {
    "scmcp": {
      "command": "scmcp",
      "args": [
        "run",
        "--run-mode",
        "tool"
      ]
    }
  }
}

If your client cannot find scmcp, put the full path printed by which scmcp in "command"

pip

TERMINAL

pip install scmcp

SCMCP