This project provides a centralized terminal interface for managing multiple AI coding agent sessions, addressing the complexity of tracking and switching between numerous running processes. It targets developers and researchers who utilize AI coding assistants like Claude Code or Gemini CLI, offering enhanced visibility, organization, and efficiency through a unified command center.

How It Works

Agent Deck is built using Go and the Bubble Tea TUI framework, with tmux as its underlying session management layer. It allows users to attach and detach Model Context Protocol (MCP) servers on-demand per project or globally, fork AI conversations for parallel experimentation, and significantly reduce memory overhead via an MCP Socket Pool that shares MCP processes across sessions. Smart status detection and fuzzy search provide real-time insights and quick access to any session.

Quick Start & Requirements

• Primary Install: curl -fsSL https://raw.githubusercontent.com/asheshgoplani/agent-deck/main/install.sh | bash
• Other Install Methods: Homebrew (brew install asheshgoplani/tap/agent-deck), Go (go install github.com/asheshgoplani/agent-deck/cmd/agent-deck@latest), From Source (git clone ... && make install).
• Prerequisites: tmux (installed by the script if needed).
• OS: macOS, Linux, Windows (WSL).
• Links: GitHub Repository, Documentation

Highlighted Details

• On-demand attachment and management of Model Context Protocol (MCP) servers for AI agents.
• Session forking capability for Claude Code, enabling parallel exploration of different AI-generated solutions.
• MCP Socket Pool feature drastically reduces memory usage by sharing MCP processes across multiple sessions.
• Intelligent status detection (running, waiting, idle, error) and fuzzy search across all managed sessions.

Maintenance & Community

The project provides links to GitHub issues and discussions for community interaction. Specific details on maintainers, sponsorships, or dedicated community channels (like Discord/Slack) are not explicitly detailed in the README.

Licensing & Compatibility

The project is released under the MIT License, permitting commercial use and integration with closed-source projects. It is compatible with macOS, Linux, and Windows via WSL.

Limitations & Caveats

Full feature integration, such as session forking, is primarily targeted at Claude Code; other tools receive status detection and organization. Windows users must utilize WSL for operation.