> 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/agents-in-the-workspace/learnings.md).

# Learnings

The agent's persistent notes on your work, which carry over from one session to the next.

**Learnings** are the agent's persistent notes on a piece of work: what it's trying to achieve, what it has assumed, the approach it chose and what it has found. They carry over between sessions, so when you come back to a notebook days later, the agent picks up where it left off instead of starting again.

Learnings are kept mainly by [Evolve Mode](/zerve-docs/agents-in-the-workspace/evolve-mode.md), and by every worker on a [superagent's](/zerve-docs/superagents/superagents.md) team.

## What Learnings contain

Learnings have two parts.

### Overview

The **overview** is the agent's strategy document for the work. It's always loaded when the agent starts working, and has these sections:

| Section                                   | What it holds                                                         |
| ----------------------------------------- | --------------------------------------------------------------------- |
| **Project Objectives**                    | What the work is trying to achieve                                    |
| **Key Assumptions**                       | What the agent is taking as given                                     |
| **Current Hypothesis & Success Criteria** | What it currently believes, and how it will know whether that's right |
| **Chosen Approach**                       | How it's going about the work                                         |
| **Next Steps**                            | What it plans to do next                                              |

### Topics

**Topics** are longer research documents on particular themes, which grow as the work goes on. For example:

* Methodology
* Methods explored, and why some were dropped
* A log of hypotheses and experiments
* Prior work and literature
* What the data looks like
* Known limitations

The agent adds to existing topics rather than creating a new one for every result, so they stay organized. Topics can link to each other, and the table of contents is generated automatically.

Each topic has a short summary, up to 280 characters, and the overview and each topic can be up to 128 KB.

## Where to find them

| Where             | How                                   |
| ----------------- | ------------------------------------- |
| **In a notebook** | Open the notebook's **Learnings** tab |
| **In a thread**   | Open the thread's **Learnings** view  |

Learnings update live while the agent works, so you can watch them change.

Learnings can cite [sources](/zerve-docs/agents-in-the-workspace/sources-and-citations.md), so you can check where a finding came from.

## Learnings across notebooks

When you give the agent another notebook as a reference, it can also read that notebook's Learnings. That's a quick way to give it the background from earlier work. See [Giving the agent context](/zerve-docs/agents-in-the-workspace/giving-the-agent-context.md).

When a superagent assigns work to a worker, it can give the worker starting learnings, so it doesn't begin from scratch. See [Memory](/zerve-docs/superagents/memory.md).

## Learnings and skills

Learnings are specific to one piece of work. To capture a way of working that should apply everywhere, such as how your team validates a new factor, use a [skill](/zerve-docs/agents-in-the-workspace/skills.md) instead.


---

# 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/agents-in-the-workspace/learnings.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.