Glossary

Learn the key terms used in Sprinto AI, including AI Playground, Ask AI, agents, entities, and more.

This glossary defines key terms used in AI Playground and Ask AI. Use it as a reference when working with AI-driven workflows in Sprinto.

---

Glossary of Terms

Term	Definition
AI Agent	A configurable, AI-powered action that performs a specific compliance task, such as summarising a policy or reviewing a vendor questionnaire.
AI Playground	A dedicated section in Sprinto where you can create, configure, and manage AI Agents.
Ask AI	A contextual assistant available in entity drawers that allows you to ask natural language questions or trigger AI Agents.
Entity	A module or record in Sprinto, such as Policy, Risk, Vendor, Staff, Incident, or Asset. AI Agents are always scoped to a specific entity.
Agent Name	The descriptive title of an AI Agent, which indicates its function. For example, Policy Summariser.
Input Fields	The attributes from an entity record that an AI Agent uses to analyse data or generate outputs.
Instructions	The AI prompt that guides an agent’s behaviour and defines the task it performs.
Action Type	The mode in which an AI Agent operates: Display Results (show inline outputs) or Update Fields (write outputs into entity fields, coming soon).
Sprinto-defined Agents	Pre-built, pre-tested agents available in the Playground. They are disabled by default and cannot be edited.
Custom Agents	AI Agents created and configured by users in the Playground to meet organisation-specific needs.
Deploy	The action of making an AI Agent active and available across relevant entity records.
Pause	The state in which a deployed AI Agent is temporarily disabled and unavailable in Ask AI.
Draft	The initial state of a newly created AI Agent before it is deployed.
Inline Results	Read-only AI outputs that are displayed in the Ask AI drawer.
Update Fields	A future capability allowing AI Agents to write their results directly into entity fields.
Session	A temporary interaction with Ask AI scoped to a specific entity record. Sessions reset when the browser is closed, refreshed, or the user logs out.
Feedback Mechanism	The thumbs up/down feature available on AI outputs that helps refine response accuracy.