Six coding agents, one menu bar: where the "waiting on you" signal actually lives

Description and use

Agent Ping is a lightweight status panel that reads local logs from multiple AI coding agents to show which are waiting for input, idle, or errored, enabling quick resumption of work.

Problem

Managing multiple AI coding agents simultaneously is hard because it is difficult to know which agent is stopped or waiting for user input.

Outcome

Agent Ping provides a lightweight, network‑free status panel that accurately reflects the state of each coding agent by reading local logs, enabling developers to quickly identify and resume sessions that are waiting for input or have errored.

Workflow

• Run multiple agents concurrently on a laptop
• Observe that windows look identical regardless of state
• Write Agent Ping to read local logs from each agent
• Parse logs to determine session state (waiting, idle, errored, etc.)
• Display sessions in a menu bar panel grouped by state
• Click a session row to return to the agent’s session

Evidence

I run six coding agents on the same laptop: Qoder, Claude Code, Codex, ZCode, WorkBuddy, OpenCode.

Source

The hard part is not throughput. It is knowing which one is stopped.

Source

Every one of those six tools writes a local log: Codex keeps rollouts, Claude Code keeps transcripts, the others keep session stores and local databases.

Source

Agent Ping, a menu bar panel that reads those logs and sorts sessions into five states, grouped, with the "waiting on you" group first — and it carries the verbatim question, not "1 pending item".

Source

The unit of value is one row: a session, a state, a way back.

Source