Dynamic HTTP request body in AI agent

Hello everyone,

I've been struggling with this for hours and would greatly appreciate some assistance.

I'm working on creating an AI agent that generates a draft invoice from a simple voice command. Using the API I'm integrated with, after an invoice is initially created (as empty), a route is triggered to add articles. These articles are provided as a structured array, where each object in the lines array represents an article. Here's an example of the request body:

{
  "lines": [
    {
      "unitPrice": "{unitPrice}",
      "quantity": "{quantity}",
      "quantityUnit": "unit",
      "designation": "{designation}",
      "type": "sale_of_goods",
      "vatCode": "FR_2000",
      "isDeliveryOfGoods": true
    }
  ]
}

As you can see, I've used placeholders ({}) for values that the model is expected to populate. The objective is to be able to say something like “Add 40 books and 10 hats to the invoice”, and have the model correctly fill in the corresponding values.

The issue I'm facing is that while the data is populated correctly, the agent ends up calling the route twice. Each call includes a single article, rather than making a single call with both articles in the same payload.

I've attempted to generate a proper JSON containing both objects using the CodeTool and then feeding it into the HTTP Request node, but I haven't been able to make it work, or even confirm if it's feasible.

Here is the link to the workflow:

• callin.io version: 1.83.2
• Database (default: SQLite): In-memory
• callin.io EXECUTIONS_PROCESS setting (default: own, main):
• Running callin.io via (Docker, npm, callin.io cloud, desktop app): Cloud
• Operating system: macOS