> 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/integrations.md).

# Integrations

The Operator Agent can interact with external systems through connected integrations. Once an app is connected, the agent can use its actions as tools — reading data, sending messages, updating records, and more.

### How Integrations Work with the Agent

The agent draws on connections from two places:

* **Project connections** - Apps connected to the project are available to the agent directly in chat. You can see and manage them from the Connections panel in the composer's + menu.
* **Skill connections** - Skills can bundle their own integration credentials, so a skill always runs with the accounts it was built with.

For example:

* **Gmail** — Read emails, send replies, search for messages
* **Business Central** — Look up orders, update records, create entries
* **Slack** — Send messages, post to channels
* **Google Sheets** — Read and write spreadsheet data
* **Salesforce** — Query and update CRM records

The agent decides when and how to use these integrations based on your request. You don't need to specify which integration to use — the agent matches your task to the available tools.

### Managing Connections in Chat

The Connections panel in the composer's + menu shows the apps available to the agent in the current conversation:

1. **Open the + menu** in the composer and choose Connections.
2. **Toggle apps on or off** for this conversation - switched-off apps won't be used by the agent until you switch them back on. The toggles are per conversation, so a new session starts with everything enabled.
3. **Connect a new account** from the same panel if the app you need isn't listed.

> 💡 The agent can also offer to connect an app mid-conversation. If your message mentions an app that isn't connected yet, the agent can walk you through authorising it right there in the chat.

### Connecting Through Skills

Skills can carry their own integrations:

1. Open the skill that needs the integration
2. Add the integration app as a component
3. Connect your account when prompted

Credentials attached to a skill are scoped to that skill - the agent uses them when running the skill, and won't borrow them for unrelated work.

> ⚠️ Each integration connection requires authentication with the external service. Make sure you have the right permissions in the target system.

### Available Integrations

Lleverage supports a wide range of integration apps. See the [External Apps Integration](/external-apps-integration.md) section for the full list of supported apps and their configuration details.

### Integration Actions in Chat

When the agent uses an integration during a session, you'll see it reflected in the chat:

* **Tool cards** show which integration action is being called
* **Narration** describes what the agent is doing with the integration
* **Results** are summarised in the agent's response

This gives you visibility into exactly how the agent is interacting with your external systems.

### Security and Credentials

* The agent only uses connections that are available to it - project connections you've enabled for the conversation, and credentials explicitly attached to its skills.
* Integration connections use **OAuth or API key authentication** depending on the service.
* Credentials are managed through the platform's secure credential store.
* See [Secrets and Global Variables](/secrets-and-global-variables.md) for more on credential management.


---

# 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/integrations.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.