Canonical: https://botbento.com/usecases/queue-coding-agent-work-that-needs-a/
Format: Markdown representation of the public HTML page.

# Queue coding-agent work that needs a human decision

[Coding](https://botbento.com/usecases/coding/) · Intermediate

Collect waiting agent sessions into one ordered review queue

Watch several coding-agent sessions at once and route anything blocked on a person into a single ordered queue, with a short summary of the pending change and the open design questions. Prepare review comments and quick comprehension checks so decisions are fast. Track what has been answered and what is still waiting as sessions move on.

## Quick answers

**What does it do?** Collect waiting agent sessions into one ordered review queue. Result: Ordered review queue digest.

**What does it need?** Tools: GitHub, Terminal, Chat. Intermediate: some setup. The prompt includes an approval step.

**How do I run it?** When something happens: Agent session needs input. Copy the prompt below into a bot in BotBento.

## How it works

1. When something happens: Agent session needs input
2. Then: Summarize change and open questions
3. Result: Ordered review queue digest

## Prompt to start with

```text
Monitor the configured coding-agent sessions and detect when one is waiting for input, a review, or a decision. For each item, record the session, the pending change, and the exact question being asked; rank items by urgency and dependency. Draft review comments and one or two questions that check the user's understanding of the change. Posting messages, sending comments back to agents, and closing or deleting queue items must wait for the user's approval. Keep a running list of answered and outstanding items.
```

## Tools it uses

Terminal, GitHub, Chat

## Works with

BotBento

## Tags

coding-agents, code-review, triage, workflow, terminal

## Source

[Show HN: Aq – an attention queue TUI for Claude Code](https://news.ycombinator.com/item?id=50452257) — Hacker News · checked 2026-11-12

Summarized in our own words. Any results described there are the authors’ own, not measured by BotBento.

Verified 2026-11-12: passed every automated gate and a second-model review (kimi-code/k3.1-pro). Automated, not human review. [How verification works](https://botbento.com/usecases/verification/)

By BotBento Editorial · Added 2026-11-12 · Updated 2026-11-12 · [Suggest a correction](https://botbento.com/contact/)

## Related use cases

- [Exporting terminal coding sessions into pull request notes](https://botbento.com/usecases/exporting-terminal-coding-sessions-into-pull-request/) — Convert raw agent commands, prompts, and file diffs into clean review documentation
- [Coordinate multi-agent coding and review pipelines](https://botbento.com/usecases/coordinate-multi-agent-coding-and-review-pipelines/) — Orchestrate specialized agents for research, implementation, and automated test runs
- [Batch processing for non-urgent codebase reviews](https://botbento.com/usecases/batch-processing-for-non-urgent-codebase-reviews/) — Package and submit deep code analysis tasks to deferred batch queues to reduce processing expenses
