SCMCP

What is SCMCP?

SCMCP is an MCP server that enables scRNA‑Seq analysis through natural language commands. It wraps common scanpy functions, providing modules for IO, preprocessing, clustering, differential expression, plotting, and cell‑cell communication analysis.

How to use SCMCP?

Install with pip install scmcp. Run locally using scmcp run and configure your MCP client with the command scmcp run. For remote access, run scmcp run --transport sse --port 8000 and connect to http://localhost:8000/sse.

Key features of SCMCP

• IO module for reading/writing scRNA‑Seq data with natural language
• Preprocessing: filtering, quality control, normalization, scaling, highly variable genes, PCA, neighbors
• Tool module for clustering and differential expression analysis
• Plotting module: violin, heatmap, dotplot
• Cell‑cell communication analysis

Use cases of SCMCP

• Perform scRNA‑Seq analysis via natural language in an AI chat client
• Integrate scanpy functions into agent or plugin workflows
• Quickly generate plots and summary statistics without writing code

FAQ from SCMCP

What can SCMCP do?

It provides natural‑language‑driven access to scanpy functions for scRNA‑Seq analysis, including reading/writing data, preprocessing, clustering, differential expression, plotting, and cell‑cell communication analysis.

Who is SCMCP for?

Anyone who wants to do scRNA‑Seq analysis with natural language, and agent developers who need to call scanpy functions from their applications.

Where can I use SCMCP?

In any AI client, plugin, or agent framework that supports MCP, such as Cherry Studio, Cline, or Agno.

How do I install and run SCMCP?

Install from PyPI with pip install scmcp, then run scmcp run locally. For remote use, run scmcp run --transport sse --port 8000 and connect via the SSE endpoint.

What transports are available?

Local (stdio) via scmcp run and remote SSE via scmcp run --transport sse --port 8000.