Most "multi-agent" tools assume you're running two or three agents in parallel. We built these because we're running thirty β and the usual advice (be careful, use branches, review diffs) stops working around agent number five.
At that point you need actual infrastructure: isolation, locking, orchestration, and a way to see what every agent is doing without opening thirty terminals.
Everything in this org is open source and MIT-licensed.
π‘οΈ GitGuardexRun many Codex & Claude agents in parallel without them overwriting each other. Isolated worktrees, file locks, PR-only merges. Auto-wires Oh My Codex, Oh My Claude, OpenSpec, and Caveman into every worktree.
Tutorial Β·
|
π ColonyLocal-first coordination substrate for fleets of coding agents. A SQLite store at |
π Codex FleetUIMulti-account Codex worker pool with tmux dashboards, a force-claim dispatcher, and auto-rescue supervisors. Steer dozens of parallel Codex sessions from one terminal. |
π authmuxMulti-identity helper for the Codex CLI. Auto-detects
|
Long-running Claude Code agent that tends the file-based memory lane in a real-time tick loop β keeps user, feedback, project, and reference memories tidy and de-duplicated across sessions. |
π§© claude-integratorClaude Code / Codex skill that scans |
πͺ¨ Recodee Rust orchestrator for running dozens of coding agents β shipping soon
π€ Handoffs Smoother flows for when an agent hits its usage limit mid-task
π§© OmX / OmC Deeper integrations with Oh My Codex & Oh My Claude
π§ Memory lanes Cross-agent, file-based memory that survives session resets
π Observability One pane of glass for every active agent, worktree, and account
react-grab Β· oh-my-codex-colony Β· open-design Β· copydesingskill Β· seomachine Β· medusa-plugins Β· codex-plugins
Issues and PRs welcome on any repo above β we read every one. If you're also building multi-agent tooling, we'd love to compare notes: open an issue on the most relevant project.