> 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/decisions-and-recommendations.md).

# Decisions and recommendations

How a superagent asks you for decisions and inputs, and how to respond.

When a superagent needs something only you can provide, such as approval for an irreversible change or access to a data source, it raises a **recommendation**. The app shows these as **decisions** waiting for you.

Superagents raise recommendations instead of guessing, using placeholder data or skipping a step that needs your review.

## Where decisions appear

| Where                     | How it appears                                                                      |
| ------------------------- | ----------------------------------------------------------------------------------- |
| **Your superagents list** | The superagent's card shows how many are waiting, for example "2 decisions waiting" |
| **Overview tab**          | A banner for each decision, labeled **Need input** or **Suggestion**                |
| **Conversations**         | The superagent may also ask in a conversation                                       |

A decision labeled **Need input** is blocking: the superagent can't make progress on one or more objectives until you respond, and the banner shows which. A decision labeled **Suggestion** is optional, and the superagent can continue without it.

## Kinds of decision

| Kind                             | What it asks                                  | What accepting does                 |
| -------------------------------- | --------------------------------------------- | ----------------------------------- |
| **Approval**                     | A yes or no question                          | Records your decision               |
| **Confirm a destructive change** | Permission for something that can't be undone | Records your decision               |
| **Choose an approach**           | A choice between two or more options          | Records your choice                 |
| **Connect data**                 | Access to a data source it needs              | Sets up the connection              |
| **Credential**                   | An API key or other secret                    | Stores the value you enter securely |
| **Add a KPI**                    | Tracking a metric it suggests                 | Adds the KPI                        |

{% hint style="info" %}
A superagent never asks you to type a secret into a conversation. When it needs a credential, you enter it in a secure form, and it's stored as a [secret](/zerve-docs/data-and-assets/constants-and-secrets.md).
{% endhint %}

## Responding to a decision

Respond from the banner on the Overview tab, using the action it offers, such as connecting a data source or opening a link. When you decide in the app, the superagent is notified straight away and continues.

You can also respond in a conversation, for example by telling the superagent "Yes, go ahead" or "I've added the connection". It records your answer itself.

If a decision is no longer relevant, tell the superagent in a conversation and it will close it. It also removes decisions that have become obsolete on its own.

{% hint style="info" %}
How often a superagent asks for decisions depends on each objective's [autonomy level](/zerve-docs/superagents/objectives-kpis-and-autonomy.md#autonomy). Set an objective to **Ask me first** to be consulted before major choices, or **Full autonomy** to let it resolve most decisions itself.
{% endhint %}


---

# 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/decisions-and-recommendations.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.