yyjeqhc/webcodex
Give cloud AI agents a real development environment on your own machines.
What it solves
WebCodex provides a way for AI agents (like ChatGPT and Claude) to interact directly with your local development environment. It eliminates the need to upload your code to a hosted workspace or a third-party service, allowing the AI to work on your files exactly where they live on your machine.
How it works
WebCodex acts as a bridge between the AI client and your local machine. It uses a Server/Runner architecture that connects to AI agents via MCP (Model Context Protocol) or HTTPS. This allows the agent to read and edit code, run terminal commands, execute tests, and use Git operations directly on your local system within configured project boundaries.
Who it’s for
Developers who want to AI-assisted coding without moving their repositories to the cloud, and those who need the AI to have access to their local toolchain (compilers, tests, and project-specific tools).
Highlights
- Local Code Residency: Your repositories stay on your local machine rather than being copied into a chat service.
- Real Toolchain Access: The agent can use your actual local compilers, formatters, and tests.
- Real-time Review: Includes a Runtime Console to guide, cancel, or reject the AI's work.
- Flexible Deployment: Offers a quick one-command
sharemode for trials, a Desktop app for long-term use, and self-hosting options for production. - Cross-Platform: Supports Linux, macOS, and Windows.
Related
- Project
- Project
- Project
- Project