I unified my terminal, SSH, Kubernetes, databases — and all my AI agents — into one open-source app

Description and use

A unified AI terminal consolidates coding sessions, SSH, Kubernetes, and databases into one interface, offering notifications, approvals, and file previews.

Problem

Managing multiple AI coding agents across terminals, SSH sessions, Kubernetes clusters, and databases becomes fragmented and hard to monitor, leading to unnoticed session completions, stalled approvals, and difficulty tracking which terminal is running what.

Outcome

The unified terminal consolidates AI session management, remote diagnostics, and ops workflows, reducing missed completions, streamlining approvals, and enabling zero‑install remote operations behind jump hosts.

Workflow

• 1. Launch the AI operations terminal.
• 2. AI agents start sessions; each appears in the inbox with live status.
• 3. When an agent requires approval (e.g., writing a file or running a command), a desktop notification is triggered.
• 4. User approves actions directly from the sidebar.
• 5. Agent’s file changes are previewed on the right pane.
• 6. Full transcripts are stored for later review.
• 7. For remote hosts, bind an AI agent to an SSH terminal and issue natural language commands; the agent runs diagnostics (df, free, top, systemctl, uptime) and requests approvals for system‑changing commands.
• 8. The AI can also fix issues after approval.
• 9. Kubernetes and database operations are accessed through integrated workspaces within the same terminal.

Evidence

I built an open-source AI operations terminal that gives every AI coding session an inbox — plus SSH, Kubernetes and databases in the same window.

Source

Desktop notifications — when a session needs you (approval, feedback), it calls you.

Source

Zero install on the remote host: the AI drives commands through the terminal you already connected, so it works behind jump hosts and bastions.

Source