Corgi: Multi-agent terminal multiplexer and management tool
A developer built an agent terminal multiplexer called Corgi to centrally manage multiple AI accounts and sessions across Codex, Claude Code, Pi, and more. It features real-time execution status streams, an F1 shortcut to jump to unread agents, quick link copying, image pasting over SSH, and CPU, memory, and GPU monitoring.
Hacker Newseiopa
CorgiCodexClaude CodePiLunaHaiku
Added 2026-10-31 · originally published 2026-10-23
Steps
- 1Prepare access to AI models or tools that support cost-effective reasoning (such as Luna, Haiku, etc.).
- 2Use Corgi's Unified Launcher to start and manage multiple AI accounts or sessions such as Codex, Claude Code, and Pi.
- 3Press the F1 key to quickly jump to the next agent with unread messages.
- 4Use the built-in monitoring to track CPU, memory, and GPU usage across workspaces while viewing real-time activity streams for each agent.
- 5Take advantage of convenient terminal interactions, such as one-click link copying, direct image pasting via cmd+v, and image thumbnail previews.
Copy-ready prompt
Please help me analyze and organize the following content, tell me the specific function of [Enter HTTP link or worktree path], and summarize the tasks currently being executed by each agent: [Paste the agent execution logs or related content here]
#AI Agent#Terminal Tool#Multiplexer#Developer Tools
Related use cases
Summarized by AI from the original source - refer to the original post for details. Copyright belongs to the original author.