Accessing External Data in a FlowHunt AI Chatbot Linked to LiveAgent
This article explains how to make data from your external systems — such as LiveAgent custom ticket fields or any other custom data source — available to a FlowHunt AI chatbot that is connected to a LiveAgent chat button.
Overview
When a FlowHunt AI chatbot is linked to a LiveAgent chat button, the chatbot operates within a FlowHunt flow. If you need your chatbot to access data from an external system (for example, to look up a customer record, retrieve a custom ticket field value, or pull information from a third-party service), this data is not automatically passed to the FlowHunt flow by LiveAgent.
To bridge this gap, you need to connect your FlowHunt chatbot to your external system using one of the two recommended approaches described below.
Recommended Approaches
Option 1: Add an API Component to Your FlowHunt Flow
FlowHunt allows you to add an API component (HTTP request block) directly within your chatbot flow. This component can call any external REST API — including the LiveAgent API — to retrieve the data your chatbot needs at runtime.
- Use the LiveAgent API GET /tickets/{ticketId} endpoint to fetch ticket details, including custom fields.
- The API component can pass the retrieved data to subsequent steps in your flow, making it available to the AI agent for generating responses.
- This approach is suitable for straightforward, single-endpoint data lookups.
Option 2: Integrate an MCP (Model Context Protocol) Server
For more complex or multi-source data retrieval scenarios, FlowHunt supports integration with an MCP (Model Context Protocol) server. An MCP server acts as a tool provider for your AI agent, enabling it to query multiple data sources dynamically during a conversation.
- Connect your MCP server to the FlowHunt AI agent within your chatbot flow.
- The AI agent can then call the appropriate tools exposed by the MCP server to fetch the required information.
- This approach is recommended for advanced use cases where the chatbot needs to access diverse or complex data at runtime.
Which Option Should I Choose?
- API component — Best for simple, well-defined data lookups from a single API endpoint.
- MCP integration — Best for complex scenarios requiring access to multiple tools or data sources during the conversation.
Important Notes
Each use case may have specific requirements depending on your external system's API structure and authentication method.
Since setup details vary per use case, it is recommended to consult the FlowHunt/LiveAgent support team for assistance with the initial configuration.
For customers with active LiveAgent and FlowHunt subscriptions, the support team offers full initial setup assistance for AI features and FlowHunt automations.