Use knowledge base

You can use a RAG knowledge base in the AI agent.

The agent gets small fragments of text (chunks) from the base for each user request and takes them into account when generating a response.

info

To access a RAG knowledge base, contact your account manager or email us at support@just-ai.com.

Connect knowledge base to AI agent

1. Prepare a knowledge base in Jay Knowledge Hub.

2. Connect the knowledge base in JAICP.

3. In the AI agent block, the Select knowledge base parameter displays the connected knowledge bases. Select one of them.

4. In the Knowledge base confidence parameter, you can also specify a value from 0.0 to 1.0. Then the agent only gets those chunks from the knowledge base that have a relevance score above this threshold.

   Recommended value: 0.8.

tip

When searching for relevant chunks, knowledge base settings are used.

Example

The example below shows the parameters of an agent who is acting as a support specialist:

• Agent ID: support_agent.

• LLM: gpt-4o.

• Agent role: Technical support specialist.

• Agent goal: Help the user solve the problem.

• Instructions: Give instructions with detailed steps.

• Select knowledge base: problems (Common problems).

  note

  In this example, a connection to the knowledge base named Common problems has already been created in JAICP. This knowledge base contains solutions to common problems reported to technical support.

• Knowledge base confidence: 0.8.

In the dialog, the agent:

1. Retrieves chunks from the knowledge base for each user request. The agent only gets those chunks for which the relevance score is higher than 0.8.

   In this example, the chunks contain solutions to the problems similar to the user’s problem.

2. Uses information from these chunks to create an instruction.

3. Sends the instructions to the user.