PastFumbler

Agent Run Monitor — Development skill for Claude Code

Development community

Desktop app for Windows, macOS and Linux that shows what your coding agents are doing, live.

How to install Agent Run Monitor

This entry records only its repository, not the path inside it, so there is no exact command to give. Open PastFumbler/agent-run-monitor and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agent Run Monitor does

Desktop app for Windows, macOS and Linux that shows what your coding agents are doing, live. Watch Claude Code, Codex, Gemini CLI, Copilot, Cursor, Aider and 6 more side by side: current step, files changed, and whether each run is done, failed, stuck or waiting for you. Read-only, local-only, no account.

Alternatives in Development

Coding Agent Session Search — Unified TUI and CLI to index and search your local coding agent session history across 11+ providers (Codex, C 708 ★
Anthropic Mythos Desktop Studio — Claude 3.5 Sonnet Desktop Client for Windows Mac Linux 2026 116 ★
Claude View — 10 Claude sessions running 104 ★

README

Agent Run Monitor

[![CI](https://github.com/PastFumbler/agent-run-monitor/actions/workflows/ci.yml/badge.svg)](https://github.com/PastFumbler/agent-run-monitor/actions/workflows/ci.yml) [![Latest release](https://img.shields.io/github/v/release/PastFumbler/agent-run-monitor?display_name=tag&sort=semver)](https://github.com/PastFumbler/agent-run-monitor/releases/latest) [![Licence: MIT](https://img.shields.io/badge/licence-MIT-blue.svg)](LICENSE) ![Platforms: Windows, macOS, Linux](https://img.shields.io/badge/platforms-Windows%20%7C%20macOS%20%7C%20Linux-lightgrey.svg)

**See what all your coding agents are doing, in one calm window. Read-only, local-only, no account.**

Agent Run Monitor is a desktop app for Windows, macOS and Linux that shows what your coding agents are doing. Claude Code, Codex, Gemini CLI, GitHub Copilot, Aider, Cursor and others save each session to a file on your computer. The app reads those files and shows every session as a **run**: what it is doing now, what it did, which files it changed, and whether it finished, failed, is waiting for you or looks stuck. It only reads. It never controls an agent, and it never changes an agent's files.

![A live run updating, stepping between runs with the arrow keys, searching with Ctrl+K and opening the multiviewer](docs/images/demo.gif)

All pictures on this page come from the app's built-in demo, which uses made-up runs.

Highlights

12 agents, one view. Claude Code, Codex, Gemini CLI, Qwen Code, Copilot in VS Code, Copilot CLI, Aider, Cursor, Cline, Roo Code, Continue and OpenCode, all shown in the same format.
Live. Runs update within a second or two of the agent writing. You see the current tool call, the latest messages and reasoning, and the files being changed.
Clear status. Running, waiting for you, idle, done, failed, stuck, or couldn't be read. Each status has its own shape and word as well as a colour.
Multiviewer. Watch up to 12 runs side by side, each tile wit