Complexipy

Provides cognitive complexity analysis for Python codebases using complexipy via MCP.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "onenightcarnival-complexipy-mcp": {
      "command": "uvx",
      "args": [
        "complexipy-mcp"
      ]
    }
  }
}

You run a dedicated MCP server that leverages the complexipy library to analyze cognitive complexity in Python codebases. It can scan entire directories or individual files, returning clean JSON data that you can feed into other tools or dashboards for quick insights and quality gating.

How to use

Start the MCP server and connect an MCP client to obtain JSON results that describe cognitive complexity at the file and function level. You can scan entire codebases or single files, and you’ll get a machine-friendly list of files that exceed your chosen complexity threshold, including per-function details and line ranges.

Typical usage patterns include: scanning a whole Python project to identify hotspots, inspecting a single file to understand the most complex functions, and sorting or filtering results to focus remediation efforts. The output is pure data, designed for automation and integration with your existing quality pipelines.

How to install

Prerequisites you need to have before starting:

A runtime tool for MCP servers, such as uvx or uv, installed on your system.

Step-by-step setup:

# Quick start: run the server directly with uvx
uvx complexipy-mcp

# Local development: run the server via the MCP runtime
uv run complexipy-mcp

# Claude Desktop (remote configuration example)
uvx complexipy-mcp

Additional notes

If you are configuring multiple run modes, you can use the following two stdio configurations to cover both quick-start and local development approaches. Each configuration runs the same MCP server but via different runtimes.

{
  "type": "stdio",
  "name": "complexipy",
  "command": "uvx",
  "args": ["complexipy-mcp"]
}

{
  "type": "stdio",
  "name": "complexipy",
  "command": "uv",
  "args": ["--directory", "/ABSOLUTE/PATH/TO/complexipy-mcp", "run", "complexipy-mcp"]
}

Available tools

scan_directory

Recursively scans a directory for Python files whose cognitive complexity exceeds a configurable threshold and returns a JSON list of files with their complexity data and per-function breakdown.

scan_file

Analyzes a single Python file to determine cognitive complexity and returns a JSON object describing the file's overall complexity and function-level details.