xintaofei/codeg

Collaborative multi-agent AI coding workspace: aggregate sessions from Claude Code, Codex, OpenCode, Pi, Grok Build, etc. Desktop app, self-hosted server, or Docker.

What it solves

Codeg is a multi-agent coding workspace that solves the problem of fragmented AI coding agents. Instead of switching between different CLI tools and silos, it aggregates multiple agents (like Claude Code, Codex, and Gemini) into a single searchable workspace where they can collaborate and work together on a a single project.

How it works

Codeg acts as a management layer over agent CLIs. It integrates various AI agents into a unified interface, allowing users to delegate tasks to different agents using @-mentions. It manages the project's codebase using git worktrees, creating separate branches for unattended tasks on a to-do board, which are then reviewed and merged by the user.

Who it’s for

Developers who use multiple AI coding agents and want a unified environment for managing their codebase, tracking tasks, and collaborating across different AI models.

Highlights

  • Unified Workspace: Aggregates sessions from various supported agents into one searchable interface with a built-in editor, diff viewer, and git client.

  • Multi-Agent Collaboration: Use @-mentions to delegate tasks to sub-agents, allowing different agents to run in parallel within a single task.

  • Office Document Integration: Live preview and editing of .docx, .xlsx, and .pptx files directly within the workspace.

  • Task Management: A to-do board where agents work on tasks in separate git worktrees, awaiting user review before merging.

  • Crosspatfrom Support: Available as a desktop app, standalone server, or Docker container, with native iOS and Android clients for remote monitoring and approval.

  • Custom Agent Support: Ability to register ACP-compatible agents from a public registry or via distribution JSON.

Related

  • Project
    multica-ai/multica

    An open-source workspace that orchestrates multiple AI coding agents, allowing users to assign them tasks like teammates and track their progress on a unified board.

  • Project
    openteams-lab/openteams

    A local-first AI desktop app that provides a unified workspace for coordinating multiple AI coding agents, allowing them to share context and work within developer-controlled workflows.

  • Project
    asheshgoplani/agent-deck

    A command center for AI coding agents that allows users to manage, monitor, and orchestrate multiple agent sessions across different projects from a single terminal.

  • Project
    openagents-org/openagents

    An open-source collaborative OS for AI agents that provides a unified workspace where multiple agents from different runtimes can share context, files, and a browser.

  • Project
    codeaholicguy/ai-devkit

    A control plane for AI coding agents that provides a unified configuration, a supervision console, and local searchable memory to coordinate multiple AI assistants.