preset-io/agor

What it solves

Agor provides a centralized, self-hosted command center for managing coding agents (such as Claude Code, Codex, and Gemini) within a team environment. It eliminates the friction of running agents in isolated terminals, providing a web-based workspace where agent work is anchored to git branches, ensuring that each agent session has its own isolated development environment and conversation history.

How it works

Agor acts as an orchestration layer that drives existing agent CLIs and SDKs rather than providing its own model. It uses a process-isolated executor to run agents on specific git branches, which serve as the primary unit of work. The system is organized around a 2D canvas (Boards) where branches are represented as cards. It also exposes an MCP (Model Context Protocol) endpoint, allowing agents to introspect and drive the Agor system itself.

Who it’s for

It is designed for software engineering teams who use multiple different coding agents and want a collaborative, governed, and isolated environment to manage their AI-assisted development workflows.

Highlights

• Branch-Centric Workflow: Every piece of work is a git branch with its own working directory, dev server, and history.
• Multi-Runtime Support: Interchangeable support for Claude Code, Codex, Gemini, OpenCode, Copilot, and Cursor.
• MCP-Native: Agents can fork, spawn, and report on their own work via the Agor MCP server.
• Long-lived AI Teammates: Persistent AI coworkers with their own searchable memory (Knowledge-base) and schedules.
• Multiplayer Collaboration: Real-time shared sessions with live cursors, comments, and a shared terminal.
• Governance: Branch-scoped RBAC, ACLs, and detailed token/cost accounting per prompt.