To create the AI agent, follow these steps:

1. Go to the "AI agents beta" tab on make.com.
2. Click the "Get started" button to create a new AI agent.
3. Add your API key by going to the OpenAI playground, clicking the settings icon, and creating a new API key. Copy the generated key and paste it into the "Connection" field.
4. Name your AI agent, such as "AI agent script writer v2".
5. Select the model you want to use, in this case, GPT-4.0.
6. In the "System prompt" section, outline the steps you want the agent to follow:
   • Search Google Sheets rows to find the information.
   • Browse Perplexity to gather the data.
   • Create the script.
7. Click "Save" to create the AI agent.

Next, we'll create the tools that the AI agent will use to complete its tasks. This includes connecting to Google Sheets to search for ideas, using Perplexity to research the topic, and creating a Google Doc to write the final script.

1. Google Sheets Search Tool:

   • Click the "Scenarios" tab and then "Create New Scenario".
   • Select "Google Sheets" and choose the "Search Rows" option.
   • Connect your Google account and select the spreadsheet and sheet containing your ideas.
   • Set the output structure to "content_idea" to pass the idea back to the AI agent.
   • Name this tool "Google Rows Search Tool" and set the scheduling to "On Demand".

2. Perplexity Research Tool:

   • Create a new scenario and select "Perplexity" with the "Chat Completion" option.
   • Add your Perplexity API key and choose the "Sonar" model.
   • Set the scenario input to "topic_idea" and the output to "topic_research".
   • Name this tool "Perplexity Research Tool v2" and set the scheduling to "On Demand".

3. Google Doc Creation Tool:

   • Create a new scenario and add the "OpenRouter" module with "Chat Completion".
   • Use the Gemini 2.5 Pro model and set the content to write a video script about the topic.
   • Add a "Google Docs" module to create a new document with the script.
   • Name this tool "Google Doc Creation Tool" and set the scheduling to "On Demand".

After creating these tools, go back to the AI agent and add them to the agent's arsenal. Now the agent has the necessary tools to search for ideas, research the topic, and create the final script.

Now that we have created the AI agent and the necessary tools, it's time to activate and run the agent. Here's how you can do it:

1. Configure the AI Agent: Go back to the "AI Agents" section and click on the agent you created earlier. Here, you can add the tools you created to the agent's arsenal. Click "Add" and select the tools you want to include - the "Google Sheets Search Tool", the "Perplexity Research Tool", and the "Google Doc Creation Tool".

2. Provide Descriptions for the Tools: For each tool, add a brief description to help the agent understand what the tool does. For example, "This tool allows you to research the topic with the web tool called Perplexity" or "This tool creates the final script using another LLM and creates a Google document."

3. Save the Agent Configuration: Once you've added all the tools, click "Save" to update the agent's configuration.

4. Trigger the Agent: There are several ways to trigger the agent to start working. One simple way is to use the "Run an AI Agent" feature. Click "Add Make AI Agent" and select the agent you created. You can then provide a system prompt, such as "Get to work", to instruct the agent to begin its tasks.

5. Monitor the Agent's Progress: The agent will now start executing the tasks you've defined, searching the Google Sheets for ideas, using Perplexity to research the topic, and then creating a Google Doc with the final script. You can monitor the agent's progress and check the output document once the process is complete.

6. Customize and Refine: If needed, you can further customize the agent's system prompt or the tools it uses to better suit your specific requirements. You can also save the agent's configuration as a blueprint and share it with the community for others to use.

By following these steps, you can activate and run your AI agent, taking advantage of the automated workflow to streamline your content creation process.

In this tutorial, we have covered the key steps to create an AI agent using the make.com platform. We started by setting up the base AI agent, defining its system prompt and instructions. Then, we created the necessary tools for the agent to perform its tasks, including searching Google Sheets for ideas, researching the topic on Perplexity, and generating a final script in a Google Doc.

The process involved connecting the agent to external APIs, configuring the tool scenarios, and ensuring the agent can only access the designated tools. Finally, we set up a trigger to run the agent and saw the final output in a new Google Doc.

This example demonstrates the power of AI agents in automating complex workflows and leveraging various AI models and tools to accomplish tasks efficiently. By following this guide, you can now create your own AI agents and customize them to suit your specific needs, taking advantage of the AI revolution.