Back to directory

siGit Code

A local coding agent that runs entirely on the developer's machine, working in any codebase through a terminal chat interface or the Agent Client Protocol for editors such as Zed and Xcode, with no API key or cloud round trip required.

siGit Code

siGit Code is a coding agent that performs inference locally rather than calling a cloud API, so it needs no API key, subscription or network round trip once its model is downloaded. It can be used in ACP mode, where an editor such as Zed or Xcode starts it over stdio, or in terminal mode as an interactive chat UI, and works in any codebase, with extra awareness of a specific company's Rust workspace layout, deploy flows and auth boundaries when used there.

It installs through cargo, Homebrew, pip, uv, npm, NuGet, winget, Scoop, or a downloaded Debian or RPM package, and supports macOS and Linux fully, with Windows supported in terminal mode only. The first run downloads a GGUF model of roughly 1 to 2 GB from Hugging Face; later runs load it from a local cache, and the macOS desktop app and CLI can share the same cache.

A hosted version, siGit Code Cloud, is available for people who would rather not run a model locally, and the project's own site also offers Git hosting built for AI workflows. It is written in Rust.

More in Coding agents

See all 793
  • The City

    github.com

    A native macOS app that shows Claude Code runs as a city, with each project as a building, each team as a floor of robots, and every animation driven by a real event from the claude CLI's stream.

  • rig

    tryrig.ai

    A small, single-binary agent runtime that points at any OpenAI-compatible model, local or hosted, reads and edits a repository's code, runs its tests and remembers what it learned, with a loop implemented in a few hundred lines.

  • Jauvex

    jauvex.reindent.com

    An Electron desktop app that puts Claude Code, Codex and Grok coding-agent sessions side by side and lets you steer them by voice, with a small model handling fast routing decisions.