Would you leave an agent running overnight if it first refused a dirty repo, a merge in progress, a missing runner, or low disk? I built those preflight refusals into Omnius. Autonomy should start…
You write the specs. It clears the board: coding agent chains plan, review, implement, verify and merge unattended on your own machine, on the GPT and Claude subscriptions you already have.
Local-first runtime for coding agents. Agents file tasks, Orbit lands them from sandboxed worktrees, and every step is audited. Works with Claude Code, Codex, Grok, Agy and Muse. No telemetry.
Your coding agents, working together as one team. Claude Code, Codex, Cursor and 13 more talk to each other, split the work on a shared board and check each other's work, each in its own git worktree. Free, open source, on your own plans and your own computer. macOS, Windows, Linux.
No long blocks of time? Run an AI team from your phone. Open-source, local: Git worktree sessions, task contracts and verification gates for Claude Code, Codex, Gemini CLI, Copilot, OpenCode, Antigravity and Command Code.