Basic Components of an Agentic AI Bot
- Getting Started
- Bot Building
- Smart Agent Chat
- Conversation Design
-
Developer Guides
Code Step Integration Static Step Integration Shopify Integration SETU Integration Exotel Integration CIBIL integration Freshdesk KMS Integration PayU Integration Zendesk Guide Integration Twilio Integration Razorpay Integration LeadSquared Integration USU(Unymira) Integration Helo(VivaConnect) Integration Salesforce KMS Integration Stripe Integration PayPal Integration CleverTap Integration Fynd Integration HubSpot Integration Magento Integration WooCommerce Integration Microsoft Dynamics 365 Integration
- Deployment
- External Agent Tool Setup
- Analytics & Reporting
- Notifications
- Commerce Plus
- Troubleshooting Guides
- Release Notes
- Agentic Bot
Table of Contents
An Agentic AI bot is primarily built using two core components:
System Prompt
Actions
System Prompt
A System Prompt is a set of instructions provided to the Large Language Model (LLM) that defines its behavior, capabilities, and decision-making process. It acts as the bot's foundational guidance and is included with every user interaction.
Since the system prompt is sent with every query, it is recommended to keep it clear, concise, and focused. A shorter prompt consumes fewer tokens, leaving more context available for the user's query and helping reduce the likelihood of hallucinations.
Actions
Actions enable an Agentic AI bot to perform tasks beyond generating text. They allow the agent to interact with external systems and execute predefined operations.
Common use cases for actions include:
Fetching real-time data from external services.
Calling APIs or third-party applications.
Invoking another prompt or a Retrieval-Augmented Generation (RAG) application.
Executing custom business logic or code.
Integrating with enterprise systems and workflows.
An Agentic AI bot can be configured with up to 128 actions, allowing it to handle complex workflows and automate a wide range of tasks.