> 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/objectives-kpis-and-autonomy.md).

# Objectives, KPIs and autonomy

Set what a superagent works toward, how success is measured and how freely it can decide.

Objectives are what a superagent works toward. KPIs measure whether it's getting there. Autonomy levels set how freely it can make decisions along the way.

## Objectives

An objective is a durable outcome, stated in one short line. For example: "Validate a weekly drawdown signal on the paper book."

A superagent can hold several objectives at once. It works best with a small, focused set: when it has more than five active objectives, it looks for ways to consolidate them.

### Stages

Each objective has a stage that shows how mature the work is:

| Stage          | Meaning                                                                                  |
| -------------- | ---------------------------------------------------------------------------------------- |
| **Discovery**  | Establishing the facts and testing whether an approach works. New objectives start here. |
| **Iteration**  | Refining an approach that shows promise.                                                 |
| **Production** | Running a validated approach as a dependable system.                                     |

Objectives are listed most mature first. See [How superagents approach work](/zerve-docs/superagents/superagents/how-superagents-approach-work.md).

### Statuses

| Status         | Meaning                       |
| -------------- | ----------------------------- |
| **Active**     | Being worked on               |
| **Achieved**   | Met                           |
| **Abandoned**  | No longer pursued             |
| **Superseded** | Replaced by another objective |

The **Objectives** card on a superagent's Overview summarizes progress, for example "1 of 2 met".

### Adding and editing objectives

You can edit objectives while reviewing the plan, and later in the superagent's [settings](/zerve-docs/superagents/settings.md). The superagent can also propose new objectives, advance their stage and mark them achieved as work progresses.

## KPIs

A KPI is a measurable success criterion. Each has:

| Field                 | Values                                                 |
| --------------------- | ------------------------------------------------------ |
| **Name**              | What's being measured, such as MRR or accuracy         |
| **Target**            | A number                                               |
| **Direction**         | **At least (≥)**, **At most (≤)** or **On target (=)** |
| **How to measure it** | Optional guidance on how to measure it                 |

A KPI can be tied to an objective, where it appears under that objective as a line such as "KR 1: Grow revenue", or be a **standing KPI** that isn't tied to any objective.

KPIs show their current value against the target, for example "120 / 100 MRR (100%)". The current value can be set by you, by the superagent itself, or reported by a worker.

{% hint style="info" %}
Good KPIs set a first milestone you can hit and measure soon, rather than an ambitious end-state number. You can raise the bar as the work matures.
{% endhint %}

## Autonomy

Each objective has its own autonomy level, which controls how freely the superagent makes decisions while pursuing it:

| Level             | What it does                                                   |
| ----------------- | -------------------------------------------------------------- |
| **Ask me first**  | Pauses before major choices                                    |
| **Balanced**      | Asks for input on major structural paths. This is the default. |
| **Full autonomy** | Proceeds without interrupting you, resolving decisions itself  |

Autonomy is set per objective, so you can give the superagent free rein on low-risk work while keeping control of anything irreversible or high-stakes.

Autonomy applies to the superagent's own decisions. Its workers don't decide on their own; they escalate to the superagent, which applies the objective's autonomy level.

{% hint style="warning" %}
Autonomy levels are different from the **Autonomous mode** switch in a superagent's settings, which controls whether the superagent keeps working through its backlog without being prompted. See [Superagent settings](/zerve-docs/superagents/settings.md).
{% 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/objectives-kpis-and-autonomy.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.