> For the complete documentation index, see [llms.txt](https://docs.zerve.ai/zerve-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.zerve.ai/zerve-docs/superagents/conversations-backlog-and-follow-ups.md).

# Conversations, backlog and follow-ups

Talk to a superagent, see what it's working on and get updates when work finishes.

You work with a superagent by talking to it. It keeps a backlog of what it's doing, and follows up with you when work you asked about is finished.

## Conversations

Your conversations with a superagent are in the **Zerve AI** panel on the right of its page. The panel opens by default, and you can resize or close it.

Each conversation is a thread. Start a new one whenever you like; the superagent brings the same objectives, backlog and [memory](/zerve-docs/superagents/memory.md) to every conversation. A new, empty conversation suggests a few starting points:

* Summarise progress against the current objectives
* What's blocked or waiting on me?
* Review recent runs and flag anything off track

In superagent conversations, you don't choose an agent mode or model. The superagent decides how to handle each request and which models its team uses, based on its [settings](/zerve-docs/superagents/settings.md).

You can also reach a superagent from outside Zerve, for example in Slack. See [Channels](/zerve-docs/superagents/channels.md).

### Sending messages while it's working

You can send a message while the superagent is in the middle of a turn. It's queued and handled when the current turn finishes.

To have a queued message handled straight away, **force-send** it. This stops only the superagent's current turn and starts a new one on your message. Work that team members are already doing carries on.

## When the superagent works

A superagent doesn't run in a constant loop. It takes a turn when something happens:

| Trigger                       | Example                                                                                                       |
| ----------------------------- | ------------------------------------------------------------------------------------------------------------- |
| **You send a message**        | In the app or through a channel                                                                               |
| **Team work finishes**        | A worker, coordinator, subagent or batch completes, fails or pauses                                           |
| **A schedule fires**          | A weekly briefing, or a check it set for itself                                                               |
| **Autonomous mode continues** | It picks up the next backlog item on its own. See [Superagent settings](/zerve-docs/superagents/settings.md). |
| **A followed item finishes**  | Work you asked to be told about is done                                                                       |

If team work finishes while the superagent is already busy, it's handled as soon as the current turn ends, so nothing is missed.

## Backlog

The backlog is the superagent's list of work items. Each has a title, details, a priority (**low**, **medium** or **high**) and a status:

| Status          | Meaning          |
| --------------- | ---------------- |
| **To do**       | Not started      |
| **In progress** | Being worked on  |
| **Done**        | Finished         |
| **Cancelled**   | No longer needed |

The **Backlog** card on the Overview tab shows a summary, such as "1 in progress · 2 open". Select **See all** to see every item. Work the team does is linked to backlog items, so the superagent can tell you how each one is going.

## Follow-ups

When a superagent tells you "I'll let you know", it means it. It links your conversation to the work item, and when the item is done or cancelled, you automatically get a message in that conversation with the outcome, even if you've moved on to other things.

For ongoing work, such as tracking a metric, the superagent can also send you interim updates in the same way.

## When the daily budget is spent

If the superagent's daily budget has run out, it still answers your messages. It can read everything, answer your questions and schedule work for when the budget resets, but it can't start new work that spends credits until then. It tells you plainly that the budget is spent and what it has queued. See [Budgets and credits](/zerve-docs/superagents/budgets-and-credits.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.zerve.ai/zerve-docs/superagents/conversations-backlog-and-follow-ups.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.