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

# How Sprinto AI Works

Sprinto AI enables you to create, deploy, and use AI-driven workflows that streamline compliance activities across the platform. It consists of two key components: **AI Playground** and **Ask AI**. Together, they allow you to build automation and interact with AI directly in the context of your compliance data.

***

### Core Concepts

* **AI actions**: An intelligent, configurable action that performs a specific compliance task, such as summarising a policy or analysing vendor questionnaires.
* **AI Playground**: A dedicated section in Sprinto where you create, configure, and manage AI actions.
* **Ask AI**: A contextual assistant available in entity drawers (for example, Policies, Risks, Vendors) where you can ask questions or trigger AI actions.
* **Display Results**: Actions that present read-only outputs inline.
* **Update Fields**: Actions that apply AI outputs directly to entity fields (planned for future releases).

***

### Workflow Overview

1. **Create actions in AI Playground**
   * Navigate to *AI Playground* from the left navigation.
   * Choose to create a new action either by:
     * **Configure with AI**: Enter a plain-text description, and Sprinto AI auto-generates the action configuration.
     * **Manual configuration**: Define the entity, input fields, and instructions manually.
   * Save the configuration to add the action to your Playground.
2. **Deploy and Manage Actions**
   * Actions can exist in different states: **Draft**, **Deployed**, or **Paused**.
   * Once you deploy an action, it becomes available for use across relevant entity records.
   * You can pause, redeploy, or edit actions as needed.
3. **Trigger Actions in Ask AI**
   * Within an entity record (for example, a policy or vendor record), click the **AI button** to open the Ask AI drawer.
   * Select an available action (either Sprinto-defined or custom).
   * Trigger the action to generate inline results or, in future versions, apply updates to entity fields.
4. **Interact with Ask AI**
   * In the same drawer, you can ask natural language questions about the active entity.
   * The AI responds in context, and you can ask follow-up questions to refine or extend the answer.
   * Session memory ensures that context is maintained as long as the session remains active.
5. **Review and Approve Outputs**
   * All AI outputs are displayed for review before being applied.
   * For display-only results, you can copy or provide feedback.
   * For update actions (future release), you must approve suggested updates before they are saved.
6. **Feedback and Refinement**
   * Each AI-generated response includes a thumbs up or thumbs down option.
   * Feedback helps improve the accuracy and usefulness of AI Actions over time.

***

### Placement in Sprinto

* **AI Playground**: Accessed directly from the left navigation.
* **Ask AI**: Accessible in the drawer of each supported entity (for example, Policies, Risks, Vendors, Staff, Incidents, Assets). The placement is consistent, but content adapts to the entity context.


---

# 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:

```
GET https://docs.sprinto.com/sprinto-ai/how-sprinto-ai-works.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.