Waishnav/devspace

What it solves

DevSpace allows AI models like ChatGPT to interact directly with your local machine's files and terminal. It removes the need to manually upload code snippets or files to a third-party platform, enabling a "Codex-style" workflow where the AI can read, edit, and execute code within your actual local project environment.

How it works

It operates as a self-hosted Model Context Protocol (MCP) server. You run the server locally on your machine and expose it via a secure tunnel (such as Cloudflare Tunnel or ngrok). Once an MCP-capable client (like ChatGPT) connects to your public URL and you approve the connection with a password, the AI can access specific project folders you've authorized. It provides the AI with a set of tools to perform file operations, search code, and run shell commands.

Who it’s for

Developers who want to use LLMs as active coding partners that can directly modify their local codebase, run tests, and manage Git workflows without leaving their chat interface.

Highlights

• Local Execution: Read, write, and edit files and run shell commands directly on your machine.
• Secure Access: Uses a password-protected approval process and user-controlled tunnels to prevent unauthorized access.
• Workspace Management: Supports isolated Git worktrees for parallel coding sessions.
• Agent Integration: Can follow project-specific instructions from AGENTS.md or CLAUDE.md and discover local agent skills.
• Cross-Platform: Supports Linux, macOS, and Windows (via Bash-compatible shells).