Cezar

Cezar manages AI coding agents on your computer or your own server. It's for developers coordinating work across repositories who want to review tasks, agent activity and code changes in one browser interface.

• Parallel coding tasks: Each task gets a separate Git worktree, with excess work queued; agents can delegate independent pieces to child tasks and receive their reports.
• Workflow checks: Combine agent steps with shell checks, assign different agents to different stages, and send failed-check errors back for another attempt.
• Review and comparison: Watch agent output and tool calls live, compare alternative attempts at the same task, and send feedback or open a draft pull request.
• Issues and automations: Launch work from GitHub, Jira or Linear issues, or trigger recurring tasks through schedules and supported tracker events.

The workspace dashboard shows running tasks and work awaiting input or review, alongside reported token usage and cost by project. An autonomous mode lets runs proceed without pausing for questions. Cezar doesn't merge changes automatically.

Cezar is free and open source under MIT. It supports Claude Code, Codex, OpenCode, Pi, GitHub Copilot CLI, Cursor Agent and Junie, using your existing agent CLI logins without a separate API key. The orchestrator runs locally or on a VPS; model access still depends on the selected agent and its service.

A server deployment can keep tasks running while your laptop is closed, and the browser interface supports reviewing work from a phone. Cezar saves its data as plain files, with editable Markdown skills and YAML workflows.