NeKo Biological Network Builder
Let your agent build and analyze biological networks from pathway databases without hand-curating every edge in scripts.
Overview
NeKo is an MCP server for the Build phase that constructs and analyzes biological networks from pathway databases via agent tools.
What is this MCP server?
- Biological network construction from pathway database sources
- Network analysis workflows exposed as MCP tools
- Stdio MCP via uvx mcp-biomodelling-servers
- Entry point mcp-neko-server in the NeKo repo subfolder
- Version 1.0.0 aligned with MCP server schema 2025-12-11
- Server version 1.0.0
- Transport: stdio
- Runtime entry: mcp-neko-server
What problem does it solve?
Pulling pathway data into custom networks for an agent feature usually means one-off Python notebooks disconnected from your IDE assistant.
Who is it for?
Builders creating agent-assisted pathway or network-graph features grounded in public pathway databases.
Skip if: Apps with no life-sciences data model; teams that only need generic SQL or CSV analytics.
What do I get? / Deliverables
Your agent can request NeKo-backed network build and analysis steps with repeatable MCP calls alongside your repo work.
- MCP-driven biological network construction and analysis
- Stdio config for NeKo in agent MCP settings
Recommended MCP Servers
Journey fit
NeKo connects pathway data to your agent during product build when you are wiring research integrations, not during distribution or customer support. Canonical shelf is integrations because the server’s job is MCP exposure of NeKo network construction and analysis tools.
How it compares
Pathway-network construction MCP, not a Boolean simulator or spreadsheet SQL server.
Common Questions / FAQ
Who is NeKo MCP for?
Solo developers and small teams using Claude Code or Cursor to automate biological network construction from pathway databases in research or biotech tooling.
When should I use NeKo MCP?
Use it in the build phase while integrating pathway-derived networks into agents, APIs, or analysis pipelines before you ship end-user features.
How do I add NeKo to my agent?
Add a stdio server with uvx, --from mcp-biomodelling-servers, and positional mcp-neko-server as defined in server.json for io.github.marcorusc/NeKo.