Simple React AI Agent with Search Tools

Back Next

Configuring and publishing the processes

To configure and publish the processes, perform the following steps:

To publish the

Search Tool

process, click Actions
in the row that contains the process and select Publish
.

To publish the

Action Detail Extractor

process, click Actions
in the row that contains the process and select Publish
.

Open the

Prompt Gemini

process.

On the Temp Fields
tab of the Start step, the Model_LLM
field is set to gemini-1.5 pro
by default. You can optionally edit the model version. For information about changing the model version, see the Gemini documentation.

Save and publish the process.

Open the

Search Tool Workflow

process.

On the Temp Fields
tab of the Start step, the Loop_Limit
field is set to 5
by default. You can optionally edit the limit. This field defines the number of times the Search Tool subprocess can be called.

You can adjust the number of repeat queries in the Search Tool to get new information and better results. However, this will increase the number of tokens used.

On the Assignment
tab of the Create Prompt to LLM
step, the Prompt
field is configured to contain the request template for the prompt to LLM. Optionally, you can configure the LLM behavior in the

Prompt_Request

field using the Expression Editor, as shown in the following sample code:

For Prompt_Request:
<GenerateContentRequestBody>
   <contents>
      <parts>
         <text>{$temp.Prompt}</text>
      </parts>
      <role>user</role>
   </contents>
    <generationConfig>
      <maxOutputTokens>2048</maxOutputTokens>
      <temperature>0</temperature>
      <topP>1</topP>
      <topK>250</topK>
   </generationConfig>
</GenerateContentRequestBody>

For the

Prompt_Request

field, enter values for the following properties:

Property	Description
maxOutputTokens	Defines the maximum number of tokens that the model can generate in its response. Setting a limit ensures that the response is concise and fits within the desired length constraints.
temperature	Controls the randomness of the model's output. A lower value close to 0 makes the output more deterministic, while a higher value close to 1 increases randomness and creativity. For example, if temperature is set to 0.5, the model balances between deterministic and creative outputs.
topP	Determines the cumulative probability threshold for token selection. The model considers the smallest set of tokens whose cumulative probability meets or exceeds topP . For example, if topP is set to 0.1, the model considers only the top 10% most probable tokens at each step.
topK	Limits the number of the highest-probability tokens to consider during response generation. For example, if topK is set to 2, the model considers only the top 2 tokens at each step, controlling output diversity and quality.

Save and publish the process.

Alternatively, you can also publish the Search Tool Workflow
process from the assets page. To publish the process, click Actions
in the row that contains the process and select Publish
. The subprocesses will be published automatically. You can modify the process as mentioned in steps 7 and 8 before publishing the process if needed.

Using the Simple React AI Agent with Search Tools recipe

Download Guide

Watch

Comments

Cloud Application Integration Homepage