Waishnav/devspace

Minimal Coding Agent Harness on MCP for ChatGPT, Claude, Hermes, Grok Bot, OpenClaw

What it solves

DevSpace allows AI assistants like ChatGPT to interact directly with your local machine's files and terminal. It solves the problem of having to manually upload code snippets or files to a chat interface by creating a secure, self-hosted bridge between the AI and your local development environment.

How it works

It operates as a self-hosted Model Context Protocol (MCP) server. You install the CLI on your machine, define which local project folders are accessible, and expose the server via a secure tunnel (such as Cloudflare Tunnel or ngrok). Once connected and authenticated with a password, the AI can use specific tools to read, write, search, and execute shell commands within the approved workspaces.

Who it’s for

Developers who want to use ChatGPT or other MCP-capable coding agents as active partners that can edit code, run tests, and manage Git worktrees directly on their local system.

Highlights

  • Local Execution: Read, write, and edit files and run shell commands for builds and tests locally.
  • Secure Access: Uses a password-protected connection and user-controlled tunnels to ensure no data is uploaded to third parties.
  • Workspace Management: Supports isolated Git worktrees for parallel coding sessions.
  • Project Awareness: Can follow project-specific instructions found in AGENTS.md or CLAUDE.md.
  • Cross-Platform: Supports Linux, macOS, and Windows (via Bash-compatible shells).

Related

  • Project
    rebel0789/codexpro

    CodexPro is a Node.js‑based local MCP server that connects ChatGPT to explicitly allowed local code repositories. It offers searchable, language‑aware indexing, safe edit/patch tools, limited Bash execution, and hand‑off planning—all performed on the developer’s machine behind a token‑protected HTTPS tunnel. The project focuses on privacy, explicit workspace boundaries, and a reviewable change loop, making it a genuine AI‑assisted coding tool rather than a hosted SaaS service.

  • Project
    openai/tunnel-client

    tunnel-client is an OpenAI‑provided daemon that creates a secure, OpenAI‑hosted tunnel so a private MCP (Model Context Protocol) server can be reached by ChatGPT, Codex, and other OpenAI services without exposing the server to the public internet. It offers health/metrics endpoints, role‑based permissions, embedded demo modes, and a Go SDK for in‑process use, and can be installed via Homebrew, Docker, or Kubernetes.

  • Project
    wonderwhy-er/DesktopCommanderMCP

    An MCP server that enables AI assistants to search, edit files, and execute terminal commands on your local machine, with support for advanced file types and process management.

  • Project
    microsoft/azure-devops-mcp

    An MCP server that gives AI agents direct access to Azure DevOps data and tools, enabling them to manage work items, repositories, and pipelines.

  • Project
    nakasyou/local-mcp

    A Model Context Protocol (MCP) server that exposes local filesystem and command execution capabilities to AI agents with built-in sandboxing and user approval workflows.