> For the complete documentation index, see [llms.txt](https://docs.lleverage.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.lleverage.ai/operator-agent/skills.md).

# Skills

Skills are the building blocks that make the Operator Agent useful for your specific business processes. A skill packages up everything the agent needs to perform a particular task: instructions on what to do, connections to external systems, and access to relevant knowledge and data.

### What Is a Skill?

Think of a skill like an app for your agent. Just as you install apps on your phone to add capabilities, you add skills to your project to teach the agent new tasks.

A skill can include:

* **Instructions**: what the agent should do when the skill is used, how it should behave, and what output to produce.
* **Connections**: connected apps (e.g. Gmail, Business Central, Slack) with the accounts needed to use them. An account can be shared by everyone, or each operator can connect their own.
* **Workflows**: existing workflows from your project that the skill can run as part of its work.
* **Knowledge bases**: document collections the agent can search for context and reference data.
* **Data tables**: structured data the agent can query and update during execution.
* **Files**: reference documents bundled with the skill, such as templates or examples. You can view and edit them from the skill page.

Each skill also has a name, a description, and an icon so it is easy to recognise across the product.

### Where Skills Come From

* **Create your own**: build a skill on the Skills page, or ask the agent to create one during a chat. The agent proposes the skill and nothing is saved until you approve it.
* **Install from the Skill Library**: browse ready-made skills built by Lleverage and install them into your project. Installed skills become project skills you can adapt like any other.

### How Skills Work at Runtime

When you give the agent a task, it evaluates which of its available skills are relevant and loads them as needed. When a skill loads, the agent reads its instructions and carries on following them in the same turn, so there is no pause between picking the skill and doing the work. You can also invoke a skill yourself, either explicitly in the message box or from the Explore page.

The agent can:

* Use **multiple skills** in a single session
* Decide **which skills to use** based on your request
* **Chain skills together** when a task requires several capabilities

### Skill Visibility

What you can do with skills depends on your project role. Project roles are Owner, Builder, Operator, and Member:

| Role                  | What you can do                                                                  |
| --------------------- | -------------------------------------------------------------------------------- |
| **Owner / Builder**   | Create, edit, test, and publish skills, including drafts and version history     |
| **Operator / Member** | Use skills through the agent. The agent only ever runs published, enabled skills |

{% content-ref url="/pages/w8GlWwXaO54V8gcV2up8" %}
[Using Skills](/operator-agent/skills/using-skills.md)
{% endcontent-ref %}

{% content-ref url="/pages/HU86KkoN2CJcfOizUE0N" %}
[Creating Skills](/operator-agent/skills/creating-skills.md)
{% endcontent-ref %}


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.lleverage.ai/operator-agent/skills.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.