> 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/get-started/faq.md).

# FAQ

Welcome to the Lleverage FAQ! Here you'll find answers to the most common questions about using the platform. If you don't see your question answered here, you can always reach out to support through the help menu in the app.

<details>

<summary><strong>What is Lleverage?</strong></summary>

Lleverage is an AI automation platform for operational work. It pairs an AI coworker (the agent, called Llev) with visual workflows, so you can hand ad-hoc tasks to the agent and turn repeatable processes into step-by-step workflows that combine logic, AI, and system integrations - without needing developers.

</details>

<details>

<summary><strong>What is the agent?</strong></summary>

The agent is an AI coworker called Llev that lives inside your project. You chat with it in natural language: give it tasks, upload files, and let it work with your connected apps, tables, and knowledge bases. It remembers context between sessions and can save repeatable ways of working as skills. New projects open on the agent by default.

</details>

<details>

<summary><strong>What are skills?</strong></summary>

Skills are reusable task packages the agent can run - a saved way of doing a job, combining instructions, integrations, and knowledge. When the agent proposes a new skill, you review and approve it before it's saved, so you stay in control of how it works. You can also pick ready-made skills from the Skill Library instead of starting from scratch.

</details>

<details>

<summary><strong>How do I get started with Lleverage?</strong></summary>

Open a new project and you'll land in a chat with Llev, your AI coworker - just describe what you need. If you'd rather build a workflow, open the visual canvas and build it step by step. The Workflow Copilot, which creates workflows from a description, is temporarily unavailable while it is being improved.

</details>

<details>

<summary><strong>Is Lleverage secure?</strong></summary>

Yes, security is a core priority at Lleverage. We are SOC 2 Type 2 certified, GDPR compliant, and ISO 27001 certified. We implement enterprise-grade end-to-end security, secure authentication options, and API token management to ensure your data and workflows are protected.

</details>

<details>

<summary><strong>Are my projects and workflows private?</strong></summary>

Yes, all your projects and workflows are private by default. Only people you explicitly invite to your organisation and projects can access them. You control who can view or edit your workflows through the permission system.

</details>

<details>

<summary><strong>How do I create a workflow?</strong></summary>

To create a workflow, choose how it starts (a form, a chat, an API call, a schedule, or an integration event), then use the visual canvas to add actions, connect them together, and configure each step. The Copilot, which generates workflows from a natural-language description, is temporarily unavailable while it is being improved.

</details>

<details>

<summary><strong>Can I connect Lleverage to other applications?</strong></summary>

Yes. Native integrations - including Gmail, Outlook, Slack, Microsoft Teams, SharePoint, Google Drive, and ERP systems such as Business Central and SAP S/4HANA - are connected in Settings under **Connections**, and thousands more apps are available through our app integrations. In the workflow editor, connected apps appear in the **Integrations** category when you add a step, and the agent can connect apps for you as part of a conversation. The Workflow Copilot, which can also connect apps, is temporarily unavailable while it is being improved.

</details>

<details>

<summary><strong>Where do approvals and human input requests appear?</strong></summary>

In the Request Inbox. When the agent or a workflow needs a person to approve something or fill in missing information, a request appears there and the run pauses until it's handled. Requests can be assigned and reassigned, carry notes, and have priorities and SLAs, so the right person deals with the right thing on time.

</details>

<details>

<summary><strong>How do I pass data between nodes in my workflow?</strong></summary>

Click on an input field where you want to use data from a previous step, then click the @ button in the field. This opens a picker where you can select the variable you need. For advanced users, you can also use the {{ }} method to access specific properties.

</details>

<details>

<summary><strong>Can I test my workflow before publishing it?</strong></summary>

Yes, use the Run Panel in the workflow editor to test your workflow with sample inputs and see the results before publishing it for real use.

</details>

<details>

<summary><strong>How do I fix errors in my workflow?</strong></summary>

Steps with configuration issues are highlighted on the canvas. Open the step to see which fields need attention, and check the validation messages in the editor for details about what's missing or misconfigured.

</details>

<details>

<summary><strong>How do I set up a loop in my workflow?</strong></summary>

Add a **For Each** node from the Controls section when adding a step, connect your input data (typically a list), and add the actions to perform on each item inside the loop. The workflow processes each item in turn. There are also **Repeat** and **Do While** nodes for loops that run a set number of times or until a condition is met.

</details>

<details>

<summary><strong>How do I publish my workflow?</strong></summary>

Click the **Publish** button in the workflow editor. The publish options depend on how your workflow starts - app links for forms and chats, or API endpoints for API call triggers.

</details>

<details>

<summary><strong>How do I share my workflow with others?</strong></summary>

First, add the person to your organisation and to the specific project containing the workflow, with a role that matches what they need to do. Then you can share the published workflow URL with them.

</details>

<details>

<summary><strong>Can I embed my Lleverage workflow in my website?</strong></summary>

Yes, for form and chat workflows you can embed your workflow in a website using the iframe code provided when you publish. This works with WordPress and other website platforms through custom HTML blocks.

</details>

<details>

<summary><strong>How do I control who can access my workflows?</strong></summary>

Lleverage uses a permission system with roles at both organisation and project level. At the organisation level, the roles are Owner, Builder, Operator, and Member; at the project level, Project Admin and Project Member. See [Roles and Permissions](/roles-and-permissions.md) for what each role can do.

</details>

<details>

<summary><strong>How do I stay on top of what my workflows are doing?</strong></summary>

The Activity page shows the run history for your project, and the notification bell in the sidebar collects updates that need your attention. Your project Dashboard also gives you an overview of how things are running.

</details>

<details>

<summary><strong>Which AI models does Lleverage support?</strong></summary>

Lleverage provides a curated selection of state-of-the-art AI models from leading providers including OpenAI, Anthropic, Google, Together AI, and Mistral. The available models are kept up to date as providers release new ones.

</details>

<details>

<summary><strong>How do I choose the right AI model for my workflow?</strong></summary>

Consider what the step needs to do. For simple, high-volume tasks, pick a faster, lighter model; for complex reasoning or nuanced writing, pick a more powerful one. You can test different models in the workflow editor and compare the results before settling on one.

</details>

<details>

<summary><strong>Can I use my own API keys for AI models?</strong></summary>

Yes, you can add your own API keys in Settings under **AI Models**. This allows you to use your existing subscriptions with providers like OpenAI and Anthropic.

</details>

<details>

<summary><strong>How do I set tokens for AI models?</strong></summary>

You can adjust token settings in the advanced settings of prompt steps. Open a prompt step, go to the Prompt Studio, then open the settings next to the model selection to find the advanced options, including tokens.

</details>

<details>

<summary><strong>What should I do if my workflow isn't working as expected?</strong></summary>

Check for validation errors in the workflow editor, review the run details in Activity for information about what happened during execution, and use the Run Panel to test changes. If you're still having issues, contact our support team.

</details>

<details>

<summary><strong>How do I report a bug or request a feature?</strong></summary>

Use the chat support in the help menu, or choose **Request Feature** from the user menu to send a suggestion to our product team.

</details>

<details>

<summary><strong>Can I get help with building complex workflows?</strong></summary>

Yes, our support team can help with complex issues. You can also request a personalised demo at <https://www.lleverage.ai/book-a-demo> for guidance and to talk about building for specific use cases.

</details>

<details>

<summary><strong>How do I set up environment variables or secrets?</strong></summary>

Go to Settings and open **Secrets** or **Variables**. Secrets securely store sensitive values like API keys and connection strings; variables hold reusable configuration values your workflows can reference.

</details>

<details>

<summary><strong>Can I create a knowledge base from my documents?</strong></summary>

Yes, Lleverage allows you to create knowledge bases from documents, websites, and other text sources. Both the agent and your workflows can use them to give context-aware answers based on your own data.

</details>

<details>

<summary><strong>Is it possible to schedule workflows to run automatically?</strong></summary>

Yes, you can use a schedule trigger to run workflows at predetermined times or intervals. This is ideal for recurring tasks like daily reports, periodic data processing, or routine system checks.

</details>

<details>

<summary><strong>Can I connect to databases in my workflows?</strong></summary>

Yes, Lleverage provides database connection capabilities. You can connect to databases like MySQL and PostgreSQL to query and manipulate data as part of your workflows.

</details>

<details>

<summary><strong>How can I monitor the performance of my workflows?</strong></summary>

Lleverage automatically logs all workflow executions. Open the Activity page to view detailed information about each run, including inputs, outputs, and execution times.

</details>

<details>

<summary><strong>How do I get workflow outputs as CSV or other formats?</strong></summary>

You can configure your workflow to format the output as CSV or other formats using the appropriate formatting steps. For downloadable files, use actions that can deliver files via email or store them in connected storage services.

</details>

<details>

<summary><strong>Can I connect my email account to workflows?</strong></summary>

Yes, you can connect email accounts like Gmail and Outlook to your workflows. This allows you to create automations that can read emails, send responses, or process email attachments.

</details>


---

# 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/get-started/faq.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.