graykode/abtop

Like htop, but for AI coding agents. Monitor Claude Code & Codex CLI sessions, tokens, context window, rate limits, and ports in real-time.

What it solves

abtop provides a centralized monitoring dashboard for users running multiple AI coding agents simultaneously. It solves the problem of tracking resource usage, rate limits, and orphaned processes across different agent sessions (such as Claude Code, Codex CLI, and OpenCode) without having to switch between multiple terminal windows.

How it works

It is a read-only Terminal User Interface (TUI) that discovers active agent sessions by scanning local process states and configuration files. It collects metadata—such as token usage, context window percentages, and open ports—directly from the local system. It can also integrate with terminal multiplexers like tmux, cmux, and iTerm2 to allow users to jump directly from the dashboard to the specific terminal running a selected agent.

Who it’s for

Developers who use AI coding agents across multiple projects and need a real-time overview of their agent's status, quota consumption, and system impact.

Highlights

  • Multi-Agent Support: Monitors Claude Code, Codex CLI, and OpenCode in one view.
  • Resource Tracking: Real-time visibility into token usage, context window fill levels, and rate limits.
  • Process Management: Detects orphaned ports spawned by agents and allows killing sessions or ports directly from the TUI.
  • Terminal Integration: Supports jumping to active agent sessions via Enter.
  • Extensible: Provides a JSON snapshot mode and a library crate for building custom dashboards.

Related

  • Project
    IvanWng97/pixtuoid

    pixtuoid is a Rust‑based terminal UI that visualises multiple AI coding agents as animated pixel‑art coworkers in a multi‑floor office. It supports many agent CLIs (Claude Code, Codex, etc.), shows token usage, tool activity, and offers themes, weather, pets, and a lofi soundtrack—all locally with no telemetry.

  • Project
    reyamira/models

    A TUI and CLI tool for browsing AI models, comparing performance benchmarks, tracking AI coding agents, and monitoring provider service statuses.

  • Project
    T1mn/pad

    A tmux-native TUI workspace for managing multiple terminal AI agents, allowing users to preview conversation history and jump between sessions quickly.

  • Project
    agent-of-empires/agent-of-empires

    A session manager for AI coding agents on Linux and macOS that provides persistent sessions, container isolation, and a web dashboard for monitoring parallel agent workflows.

  • Project
    asheshgoplani/agent-deck

    A command center for AI coding agents that allows users to manage, orchestrate, and switch between multiple agent sessions across different projects in one terminal.