webcodex

WebCodex is a self-hosted tool runtime: a Server authenticates requests and routes them to a Runner on the machine that owns the repository. Ask your assistant to inspect a repository, modify code, run tests, use Git, or investigate a failure. You do not need to move the project into a hosted workspace; requested file content and other tool results can still be returned to the AI client.

Start using WebCodex

For a temporary one-repository trial, start with Quick Trial and webcodex share. For a persistent setup, the unified installer workflow below is under development; see the validation status before choosing an artifact. Once a validated installer is available, choose the setup that matches where your code lives:

• One computer: install WebCodex on your personal workstation, open Desktop, and follow Unified installation. This setup includes Desktop, CLI, Server, and Runner in one platform package.
• Several computers: install the same package on the machine that will host the WebCodex Server and on each machine that owns repositories. Follow Unified installation and Deployment validation.

The build pipeline defines unified Windows NSIS, macOS package, and Debian 12 / Ubuntu 22.04+ .deb installer targets for x64 and arm64. The six installer variants still require native build and installation acceptance before release. Real-machine installation, reboot persistence, GUI behavior, and upgrade have not been accepted across all three platforms; this documentation makes no claim of identical cross-platform behavior. Published files are listed in GitHub Releases. The download/ directory contains static page source only; the download-page workflow builds a Release-manifest-based GitHub Actions artifact after publication, but does not host or deploy a page. For a source preview, check out the exact feature or release revision you intend to evaluate and follow Desktop development.

For Linux machines that join an existing Server without Desktop, see Runtime installation and CLI join. For advanced self-hosting, npm/runtime archives, Docker, or historical release artifacts, see Deployment.

What can it do?

• Understand and edit code — read, search, inspect, and make guarded changes inside configured projects.
• Use the real toolchain — run commands, tests, formatters, compilers, and project-specific tooling on the machine that owns the repository.
• Work with Git — inspect status and diffs while keeping repository operations visible and reviewable.
• Handle long-running work — keep jobs observable instead of requiring one model turn to stay open indefinitely.
• Support human review — use the Runtime Console, Workflow Session evidence, Jobs, and Git/diff review without a separate task/result acceptance subsystem.

Why WebCodex?

• Your code stays on your machine. The repository does not need to be copied into the chat service.
• The agent gets a real development environment. It can use the same files, Git checkout, compilers, tests, and tools you already use.
• Work survives beyond a single request. Long-running execution and evidence remain observable through WebCodex.
• Start temporary or run it long-term. Use one-command sharing for a quick session, or connect machines to a self-hosted Server for a durable setup.

How it works

AI client
   |
   | MCP / HTTPS
   v
WebCodex Server
   |
   | authenticated Runner connection
   v
Runner on your machine
   |
   +-- repository
   +-- Git
   +-- compilers / tests / developer tools

For the internal Server/Runner architecture, protocol surfaces, and authority boundaries, see Architecture, MCP, and Authentication.

Star History

This chart tracks the upstream yyjeqhc/webcodex repository.

Platform packages

The unified installer targets Windows NSIS, macOS, and Debian 12 / Ubuntu 22.04+ .deb on x64 and arm64. The build pipeline defines these targets; the six installer variants still require native build and installation acceptance before release. See Unified installation and the validation checklist. Existing release artifacts and npm/Docker deployment paths remain documented as advanced historical/current compatibility references in Deployment.

This is the beginning of the README. Read the rest at the original. Read the full README ↗