Skip to main content

​
Creating an Agent Manually

What do you need the Logic API to do? Write a description of the task you want to accomplish. In the example of an image analysis agent, the description might be: 
"Accept an image as a URL or base64-encoded data and tell me what it is.

Include items like:
- How many people are in it?
- What is the dominant color of the image?
- What's the general mood it invokes?
- What is the gender and or age of the person in the photo?"
To provide the Logic API with context, you can also include examples of input and output.

​
Editing Your Agent

agent-editor Once you have created an agent, you can edit it to better fit your needs. This guide will walk you through the steps to edit an agent in Logic.

​
Step 1: Access the Logic dashboard

Open the Logic dashboard and log in to your account. Select the agent you want to edit from the list of agents. If you have just generated an agent from a description, it will already be opened in the agent editor.

​
Step 2: Edit your agent

You can now edit the agent to suit your specific use case. Feel free to add, remove, or modify sections as needed. For example, you might include more detail about the image analysis process or add input/output examples.

​
Step 3: Use text formatting

Using the agent editor, you can use text formatting to make your agent more readable. For example, you can use bullet points to list the items you want the Logic API to include in its analysis. You can also use headings to separate different sections of your agent. For example, you might have a section for the input and a section for the output. The text formatting options are available in the toolbar at the top of the agent editor. You can use these options to organize your agent and make it easier to read.
If you want to learn more about what makes a good agent, check out the What Makes a Good Agent? section.

​
Finishing Your Agent

agent-list-hover-menu

​
Step 4: Set a title

The title is something for you to use to refer to your agent later, so choose something descriptive. You can edit the title directly in the agent editor at the top, or by clicking the … button on an agent card in the dashboard and selecting Rename. 
Image Analysis and Description

​
Step 5: Save your changes

Your changes are saved automatically as you edit. To trigger a manual save, press ⌘S on macOS or Ctrl+S on Windows. If you’re already using the agent’s API, make sure to publish your changes so that the API reflects the updated agent.

​
Step 6: Test it out!

Once you have created your agent, you can test it out using automatically generated test cases. This will help you ensure that it behaves as expected and meets your requirements. To learn more about how to test your agent, check out the Testing Agents section.