Originally published at taskhold.com.
Short answer: A terminal alert helps only if you're at the machine running the agent. To answer from anywhere, have the agent block the task with one specific question on a shared board, reply there from any browser, and let the agent read the reply when it next checks. Taskhold does this with taskhold_block and taskhold_updates, for any MCP agent.
An agent that stops to ask something and waits in a terminal you aren't looking at costs you the whole wait, and an alert only helps if you're near that machine. This guide covers answering from somewhere else. The agent puts its question on Taskhold's live board, where a blocked task jumps to the top with a reply box, and picks up your answer the next time it checks. Try it in the live board demo.
Getting the alert in the first place is a separate fix for each tool, with a guide for Claude Code, Codex CLI, Cursor and Gemini CLI or Antigravity CLI. The Slack alerts recipe and Claude Code hooks cover other ways to get told, and seeing what your agent is doing covers the rest of the board.
Answering from somewhere other than the terminal
Terminal alerts point you at one machine. When you're away from it, or running agents in several tools, you need the question somewhere you can read and answer, and the agent needs a way to pick the answer up.
- The agent calls taskhold_block with a specific question, for example "Keep the v1 endpoints alive, or delete them in this PR?"
- The task moves to the top of Taskhold's live board under Needs you. With notifications on, the open board tab shows a desktop notification.
- You type the answer and press Send, or Send and unblock to send the task back to the queue. On a phone this is the same page.
- The agent calls taskhold_updates with its cursor between steps or at the start of the next session, reads your reply, and carries on, or takes another task while it waits.
Answering several questions at once
When several agents are waiting, or one agent has a batch of decisions, your agent can link you to Taskhold's decisions page. It lists only the open questions, each with the agent's choices, a None of these option and a notes field, and one Send answers everything you filled in. Anything you leave blank stays open. Copy answers gives you the same answers as text for an agent's chat.
Questions an agent can act on
A good blocking question names the choice and the default. Put the rule in CLAUDE.md or AGENTS.md: "Can't continue? taskhold_block with a specific question, then check taskhold_updates for the answer."
- Vague: "Need clarification on the auth changes."
- Answerable: "Sessions expire after 24h today. Change to 7 days as the ticket says, or keep 24h for admin users?"
- Answerable: "STRIPE_TEST_KEY isn't set in .env. Add it, or should I mock Stripe in these tests?"
More patterns for wording questions and progress notes are in agent messages.
Getting the alert first
Claude Code, Codex, Cursor and Gemini CLI can each alert you on the machine running them, but whether an alert reaches you depends on your terminal, window focus and the kind of prompt. The settings and hooks that fix that are in Claude Code waits for input silently, Codex CLI waits for approval silently, Cursor agent waits for input silently and Gemini CLI waits for input silently. For a phone push whenever a task gets blocked, run the ntfy script or the Pushover script.
Limits to know
- Taskhold doesn't wake a sleeping agent. The agent reads your reply the next time it calls taskhold_updates, which the snippet asks for between steps and at session start.
- Board notifications are browser notifications, shown while the live board is open in a tab. Taskhold doesn't send phone push notifications itself; the ntfy and Pushover scripts above do that from your own machine.
- Permission prompts inside Claude Code or Codex still need an answer in that tool. Taskhold carries questions about the work, not tool approvals.
FAQ
Can I answer Claude Code's question from my phone without Taskhold?
Yes, for Claude Code: start the session with Remote Control and answer in the Claude app, which can also send push notifications. Taskhold is for questions from any agent, collected on one board.
What happens to a blocked task if the session ends?
Taskhold keeps it blocked with the question. Your reply waits on the task, and the next session reads it with taskhold_updates or taskhold_get before it picks up work.
Does the agent sit idle while it waits?
That's up to your instructions. A common rule is to block the task, then take the next todo, so one question doesn't stop the whole queue.
Sources
- Claude Code docs: Claude Code on mobile, checked 2026-10-26
Related
- OpenAI Agents SDK progress updates for long runs
- How to see what Claude Code is doing, live
- Claude Code waits for input silently: the fix
- Codex stuck on "Awaiting approval": the fix
- How to hand tasks to your coding agent
- A CLAUDE.md and AGENTS.md task workflow
- Codex CLI task list over MCP: watch tasks live
- Taskhold vs Claude Code Remote Control
About Taskhold
Taskhold is a hosted task list shared by a person and their AI coding agents over MCP. Agents break tasks into steps and report progress, and the person watches a live board. $12/month or $120/year per person, 7-day trial with a card, no free plan. Made by Great Work LLC.
To try it, add the MCP server https://taskhold.com/api/mcp to your agent. It connects with no sign-up, and the first time the agent uses Taskhold it gives you a link to start the trial.
- Pricing: https://taskhold.com/#pricing
Top comments (0)