Mcp Numpy
Let your agent create and manipulate NumPy arrays and numeric ops while you build dashboards, ML snippets, or data APIs.
Overview
io.github.daedalus/mcp-numpy is a MCP server for the Build phase that exposes NumPy array and numeric operations to your agent via stdio.
What is this MCP server?
- Wraps NumPy functionality as MCP tools for coding agents
- stdio MCP transport for local agent clients
- PyPI distribution as mcp-numpy version 0.1.0
- Fits Python-centric solo stacks without custom REST wrappers
- Pairs with other daedalus scientific MCP servers for richer math workflows
- Version 0.1.0 in MCP registry schema
- Transport type: stdio
- registryType: pypi, identifier mcp-numpy
What problem does it solve?
Agents often draft NumPy code that drifts from your installed version or skips edge cases unless they can call the library through MCP.
Who is it for?
Solo builders shipping Python data features, small ML utilities, or scientific calculators inside agent-driven workflows.
Skip if: No-Python shops or builders who only need static SQL reports without in-session array computation.
What do I get? / Deliverables
After registration, the agent can drive real NumPy operations as tools while you implement analytics or numeric backend features.
- stdio MCP server wiring NumPy into agent tool calls
- Reusable numeric/array operations during feature implementation
- Less friction between generated Python and executed array results
Recommended MCP Servers
Journey fit
Build is the primary journey phase because NumPy access is needed while writing backend logic, notebooks-as-code, or analytics features. Integrations captures MCP bridges that connect the agent to established Python data libraries.
How it compares
Local NumPy bridge over MCP, not a managed warehouse connector or a Cursor-only skill file.
Common Questions / FAQ
Who is mcp-numpy for?
Developers using MCP-enabled agents who write Python numeric and analytics code and want NumPy available as explicit tools.
When should I use mcp-numpy?
Use it during Build when you are implementing arrays, linear algebra snippets, or numeric transforms and want the agent to execute against NumPy.
How do I add mcp-numpy to my agent?
Install mcp-numpy from PyPI, configure a stdio MCP server entry in your client, and ensure the same Python env has NumPy available to the server process.