Agent Quick Start
A step by step guide on how to build your first Agent in Fetch Hive.
Last updated
Was this helpful?
A step by step guide on how to build your first Agent in Fetch Hive.
Last updated
Was this helpful?
Agents streamline complex processes by breaking them into manageable steps. By leveraging powerful tools like Google Search, vector database queries, and contextual history, you can build intelligent assistants without writing a single line of code.
In this guide, we'll walk through creating your first agent in Fetch Hive, configuring its capabilities, and making it ready for integration with your business applications.
To create an agent, navigate to the Agents section in your Fetch Hive dashboard. Click the "Create agent" button in the top right corner to open the setup form. Here you'll need to:
Set your agent name (e.g., "Our First Agent")
Add a description
Select a model (such as GPT-4o)
Choose a dataset (knowledge base)
Configure the instruction prompt
The instruction prompt serves as a guideline for how your agent responds, including its role, language preferences, and personality traits. This can be as detailed or minimal as your use case requires.
Fetch Hive agents come with powerful capabilities:
When enabled, your agent can search the web for the most recent information before responding to queries. This ensures your users receive up-to-date answers on topics like "recent reports about OpenAI."
This tool allows your agent to query your custom knowledge base containing:
Website content
PDF and document extracts
Custom text blocks
Perfect for support chatbots, e-commerce assistants, and any business requiring responses based on specific organizational knowledge.
This feature enables your agent to remember previous messages in the conversation, building rapport and providing coherent responses that acknowledge earlier exchanges.
The Activity Logs section provides complete transparency into your agent's operations:
View all user-agent message exchanges
Examine search queries and results
Review response generation
Monitor usage costs
This information helps you optimize your agents by revealing opportunities to improve your vector database or refine instruction prompts.
Once you've tested and refined your agent in the dashboard, it's ready for integration with your business applications via the Fetch Hive API. You can:
Create, retrieve, and delete agents
Manage chats and messages
Integrate with your existing systems through simple API requests
This workflow allows you to thoroughly test your agents before deploying them to your users, ensuring they provide the experience you intend.