These steps will guide you through the process of deploying an agent on Headless Agents. Before you get started, make sure you sign up for a Headless Agents account at headlessagents.ai.
Navigate from the dashboard to the deploy module.

Select your deployment type

Select the type of agent you want to deploy:

Platform Deployment

Standard Deployment

Platform Deployment Quickstart

Deploy an agent built on another platform or server using HTTP requests. Our AI-assisted platform deployment module makes it easy to deploy from any API-based service.

1. Navigate to the deploy module and select platform from the deployment selector.

Platform Selection

2. Copy the HTTP request from your platform’s documentation or API reference.

For example:
curl -X POST 'https://api.yourplatform.com/v1/agent/call' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"input": {"prompt": "Hello, world!", "style": "classic"}}'
You can leave the YOUR_API_KEY placeholder in the HTTP request. We’ll use the secrets manager to handle the authentication during the setup process.
Ensure you have the fields configured the way you want them to be called by the agent, users calling the agent will only be able to pass through a single placeholder field.

In this example, prompt is likley the field that you will want to map the user input to, in that case the style field will be continue to be set to classic.

This will be configured during the setup process.

3. Begin deployment with the Deployment Assistant

1

Paste API details

Paste the HTTP request from your platform into the deployment assistant. The assistant will automatically parse all necessary requirements to call your agent headlessly.
curl -X POST 'https://api.yourplatform.com/v1/agent/call' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"input": {"prompt": "Hello, world!", "style": "classic"}}'
2

Import an API key from the Secrets Manager

If your request requires authentication, the deployment assistant will ask you to import an API key from the Secrets Manager.Import API Key
3

Test the connection

The deployment assistant will ask for permission to send a test request to verify the connection. This ensures your API key and configuration are working correctly.Test ConnectionYou’ll see a success message when the test is complete, confirming that your external agent is accessible.
4

Configure request placeholder mapping

After a successful test, the deployment assistant will ask you to map the user’s input to the correct parameter in your API request.Configure Request Placeholder MappingThis is where you will specify where the user’s input will be passed through.When the user sends a prompt to your agent, the prompt will be replace the
User Input field in your API request.
Click the field you want to replace or drag the User Input placeholder to the correct value position in your API request.Configure Request Placeholder Mapping
5

Test user input

Next, you’ll be asked to run another test, but this time with a sample user input passing through.Test User InputEnter a sample prompt to replace the User Input placeholder in the test.Test User InputThe deployment assistant will send another test to ensure everything is working as expected.
6

Configure response handling

After a successful test, the deployment assistant will ask you to specify what part of the API response you want to return to the user.Configure Response HandlingThis ensures users receive only the relevant information from your agent.Click the field you want to return to the user in the response field of the agents response.You can return a single field:Single Field ResponseOr select multiple fields to return:Multi Field Response
7

Add a name and description

Finally, add a name and description for your agent.Enter NameEnter Description
8

Deploy your agent

Your agent is now ready to be deployed. Click Deploy!Deploy RequestDeploy SuccessGo to the dashboard to view your deployed agent.Deployed Agent
Congratulations! You’ve deployed your platform agent. Your agent is now available anywhere!

Standard Deployment Quickstart

We reccomend you deploy one of our pre-built templates to get started. You can deploy the agent in just three clicks from the deploy module. 1. Select a template from the dropdown menu. 2. Test the function provided to the agent, this function will serve as a tool your agent can call if it chooses to.

Function Notes

3. When the function test is successful, click ‘Deploy Agent’!
Note: If the function test is not successful, you can edit the function and try again.
Congratulations! You’ve deployed your first agent. Head over to the Call Agent Guide to learn how to call your agent. To learn more, read the full guide: Standard Deployment Guide.