Skip to main content
Glama

πŸš€ MCP Local Dev

Let AI handle your local development environments while you focus on building amazing things!

✨ What's This?

A local development environment manager that lets LLMs configure and manage dev environments for you. Built for AI assistants to handle environment setup, dependency management, and testing automatically.

Related MCP server: MCP Build Environment Service

πŸƒ Quick Start

  1. Install Claude Desktop from the MCP quickstart guide

  2. Add the following to your Claude Desktop config:

{
  "servers": {
     "local_dev": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/txbm/mcp-local-dev@main",
        "mcp-local-dev"
      ]
    }
  }
}
  1. Point Claude at any GitHub repository and ask it to set up a dev environment!

  2. Have it run the tests and report coverage!

  3. Have a discussion, poke around or clean it up if you're done!

🎯 Core Features

Test Runners

  • πŸ§ͺ pytest with coverage reporting

  • ⚑️ Vitest with V8 coverage

  • πŸƒ Jest with detailed coverage metrics

  • πŸ”¬ unittest with coverage support

Runtime Support

  • 🐍 Python with UV package management

  • πŸ“¦ Node.js with NPM

  • ⚑️ Bun runtime and package manager

Environment Management

  • πŸ—οΈ Automatic runtime detection

  • πŸ“¦ Smart package manager selection

  • πŸ”’ Sandboxed environments

  • 🧹 Automatic cleanup

  • πŸ”„ GitHub repository support

  • πŸ“‚ Local project support

Developer Experience

  • 🎯 Zero configuration needed

  • πŸ“Š Structured JSON logging

  • πŸ” Detailed test coverage metrics

  • πŸ›‘οΈ Isolated environments per project

πŸ’« Under the Hood

  • MCP Server Spec: Full compliance with comprehensive test coverage

  • Path Isolation: Each environment is neatly contained

  • System Integration: Uses your installed runtimes (Python, Node.js, Bun)

  • Package Management: Automatically selects fastest available package manager for each runtime

  • Network Access: Full connectivity for package management

  • Process Handling: Native system processes for maximum speed

🌟 Behind the Scenes

Development involved rigorous testing across multiple models:

  • πŸ† Claude 3.5 Sonnet: Crushed it

  • πŸ’ͺ DeepSeek V3: Strong performer

  • πŸ‘Ž O1: Not great, Bob

πŸš€ Key Takeaways

This project demonstrates the incredible potential of AI-assisted development:

  • πŸƒβ€β™‚οΈ Lightning fast prototyping

  • 🎯 That last 15% is still where the real work happens

  • πŸ“š Great example of real-world AI development patterns

πŸ’­ A Note on AI & Development

As someone who's spent years in software development, what's exciting about this project isn't just automation - it's the shift in how we interact with development environments. The value isn't in replacing human developers, but in reducing cognitive overhead. When AI handles environment setup and maintenance, developers can focus more on architecture and design decisions.

This project demonstrates that AI isn't just about generating code - it's about managing complexity. By handling the mechanical aspects of development environment setup, we free up mental bandwidth for the creative and architectural challenges that truly need human insight.

πŸ™ Big Thanks To

  • UV - Speed demon Python package installer

  • Aider - Your AI pair programming buddy

  • Anthropic - For Claude's assistance in development

  • Helix Editor - Modal editing at its finest

πŸ“„ License

MIT

Available Tools

4 tools
local_dev_cleanupC

Clean up a local development environment

Inputs: env_id (required).

local_dev_from_filesystemC

Create a new local development environment from a filesystem path

Inputs: path (required).

local_dev_from_githubC

Create a new local development environment from a GitHub repository

Inputs: github_url (required).

local_dev_run_testsC

Auto-detect and run tests in a local development environment

Inputs: env_id (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.4/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: cleanup, creation from filesystem, creation from GitHub, and running tests. There is no overlap in functionality, making it easy for an agent to select the correct tool for each specific task without confusion.

    Naming Consistency5/5

    All tool names follow a consistent 'local_dev_' prefix with descriptive suffixes (cleanup, from_filesystem, from_github, run_tests). This snake_case pattern is uniform throughout, making the tools predictable and easy to understand at a glance.

    Tool Count4/5

    With 4 tools, the count is reasonable for managing local development environments, covering key operations like setup, cleanup, and testing. It might be slightly lean for broader development tasks, but it's well-scoped for the apparent purpose without being overwhelming.

    Completeness4/5

    The toolset covers core lifecycle operations: creation (from two sources), cleanup, and testing. Minor gaps might include updating or monitoring environments, but the existing tools provide a solid foundation for basic local development workflows without significant dead ends.

    Related MCP Connectors

    Related MCP Servers

    • -
      license
      Not graded
      quality
      Not graded
      maintenance
      Enables AI assistants to automatically analyze GitHub repositories and set up development environments by detecting tech stacks, installing dependencies, and verifying project builds. Provides safe tools for repository cloning, file system operations, package installation, and build verification through an allowlisted command system.
      -
    • A
      license
      A
      quality
      D
      maintenance
      Provides secure access to containerized build environments for software projects, enabling AI assistants to execute builds, run tests, manage git operations, and inspect build artifacts without requiring local installation of dependencies.
      6
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      Provides an isolated workspace for testing candidate code, runs tests, and returns deterministic pass/fail verdicts. Enables automated grading of software engineering solutions by ensuring reproducible test runs.
      5
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      Enables AI coding agents to run fast, sandboxed pre-flight validationβ€”tests, type-checking, linting, security audits, Git safety scans, and fix suggestionsβ€”before committing or pushing code, with instant caching and optional Docker isolation.
      10
      5 npm
      MIT