> 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/getting-started/quickstart-evolve.md).

# Quickstart: Evolve Mode in a notebook

Run your first task with Evolve Mode, Zerve's most capable agent mode.

In this quickstart, you'll give Evolve Mode a task, watch it plan and delegate the work, and review what it produces. Evolve Mode is Zerve's most capable agent mode, and the same agent that powers every superagent's workers.

## 1. Start from the Workspace home page

1. Open the product switcher and choose **Workspace**.
2. On the home page, find the agent composer.
3. Make sure **Evolve Mode** is selected. New conversations use it by default. To switch modes, use the mode selector in the composer or press **Shift+Tab**.

## 2. Check the Evolve Mode settings

Evolve Mode has two settings worth checking before you start:

| Setting                 | What it does                                                                                                                                                |
| ----------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Credit cap**          | The most this run can spend. The default is 50 credits. When the run reaches the cap, it pauses so you can resume or stop it.                               |
| **When it needs input** | Whether the agent asks and waits, asks and continues after five minutes, or uses its best judgement. The default is to ask and continue after five minutes. |

Some questions always wait for you, whatever the setting, such as when the agent needs a credential or wants to do something that can't be undone. See [When the agent needs input](/zerve-docs/agents-in-the-workspace/evolve-mode/when-the-agent-needs-input.md).

## 3. Describe your task

Write what you want in plain language, then send it. For example:

> Build a proxy for US inflation from public data, compare it to the official CPI over the last ten years, and explain where they diverge.

You can also give the agent more context:

* **Upload Files** to attach data or documents.
* **Use Skills** to apply your team's best practices.
* **Reference Notebooks** to let the agent read existing notebooks.

See [Giving the agent context](/zerve-docs/agents-in-the-workspace/giving-the-agent-context.md).

## 4. Watch it work

Zerve creates a notebook for the task and the agent gets to work. As it goes, you'll see:

* **Its plan**, including the specialists it delegates to, such as coding, research or peer review agents, and their progress.
* **Its reasoning**, as a timeline of thinking, hypotheses, decisions, runs, observations and blockers.
* **The notebook**, as the agent adds and runs blocks.

You can send messages while it works. They're queued and handled in order, and if a new message changes an earlier instruction, the agent follows the latest one.

## 5. Review the results

When the agent finishes, it replies with what it found. Everything behind the answer is in the notebook, so you can inspect it:

* **Blocks** hold the code, and their outputs and variables hold the results.
* **Learnings** hold the agent's persistent notes: the objective, key assumptions, the approach it chose and research topics. See [Learnings](/zerve-docs/agents-in-the-workspace/learnings.md).
* **Sources** hold the research references it used and cited. See [Sources and citations](/zerve-docs/agents-in-the-workspace/sources-and-citations.md).
* Any **documents** it wrote appear as file outputs, with a version history.

## Next steps

* [Evolve Mode](/zerve-docs/agents-in-the-workspace/evolve-mode.md)
* [Agent modes](/zerve-docs/agents-in-the-workspace/agent-modes.md), to compare Evolve Mode with Ask, Plan and Build
* [Quickstart: your first superagent](/zerve-docs/getting-started/quickstart-superagent.md), to have a team of Evolve Mode agents work toward a goal for you


---

# 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/getting-started/quickstart-evolve.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.