Back to directory

Fusion CLI

A local, conversational control plane that lets several AI coding models propose changes to a repository, while Fusion itself verifies and applies every change only after the user approves its exact bytes.

Fusion CLI

Fusion classifies each line of a conversation itself, with no model involved: talking, explaining, planning and analysis are read-only turns, and only an explicit change request can lead to a write, which then goes through a verified build-and-delivery route. A build shows its plan, including the task, exact files and verification, before it runs, and a delivery shows a summary of files, verification result, review result and a manifest digest before it can be applied; nothing is committed or pushed automatically.

For small questions it uses one model; for a broad analysis of a large project, a Lead model decides whether to answer directly or delegate bounded investigations to parallel Explorer models, with a separate Reviewer model giving a second opinion, all capped by a budget the host policy enforces.

Its primary host is Windows 11, with writer verification running inside Docker Linux containers; unrestricted autonomous write mode is not enabled.

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.