MCP Code Checker

Example Use Cases

• 1

  Run static analysis (pylint) on a codebase and obtain filtered, LLM-friendly summaries of problems and suggested fixes.

• 2

  Execute test suites (pytest) in a controlled project directory and receive detailed failure reports and prompts for remediation.

• 3

  Perform type checking (mypy) across specified project directories and combine results with lint and test outputs for a holistic code-quality assessment.

Description

This server provides tools to run pylint, pytest, and mypy against a project directory and to generate smart prompts for LLMs to explain issues and suggest fixes. It is configurable via CLI parameters (target directories, disable codes, pytest markers, mypy strictness, etc.) and supports running all checks together for a comprehensive report. The server emphasizes security by containing operations within the project directory, offering structured JSON logging, and optional virtualenv/python configuration. It is intended to integrate with MCP-compatible clients (Claude Desktop, VSCode, mcp-config) to enable AI-assisted code quality workflows.