3  Testing API Connection

Many of the interactive portions of this workshop depend on successfully being able to connect to a generative AI model via API (we will discuss API keys in more depth later). For the purposes of this workshop I’ve set up an API key through Anthropic for you all to use (see here).

The following is a very basic function for you to test that you can connect to Anthropic’s Claude Sonnet 4.5 model. As we progress through the workshop we’ll use a slightly different function that gives you more control over some of the model generation parameters.

To use this function you must have a .Renviron file in your working directory, as the function pulls the API key from that environment file. I will securely distribute this file to workshop participants (contact me if you have not yet received this file). If you’ve just added the .Renviron file to your directory, you’ll need to refresh your R session in order for it to be registered in the environment. You can do this via the tabs at the top of the R Studio interface: Session –> Restart R.

3.1 call_claude function

You can download the following function by clicking the copy button (looks like a clipboard) in the code chunk displayed below. Alternatively, you can click the icon below:

📥 Download call_claude

# Required packages
library(httr)
library(jsonlite)

call_claude <- function(prompt,
                        model = "claude-sonnet-4-5-20250929") {
  
  # Get API key from environment
  # You will need to change this to your own API key after workshop
  api_key <- Sys.getenv("ANTHROPIC_API_KEY")

  # Convert text prompt to required message format
  messages <- list(list(role = "user", content = prompt))
  
  # Build request body
  request_body <- list(
    model = model,
    messages = messages,
    max_tokens = 1024 # Required; will be an argument in other functions
  )
  
  # Set up headers
  headers <- add_headers(
    "x-api-key" = api_key,
    "anthropic-version" = "2023-06-01",
    "content-type" = "application/json"
  )
  
  # Make the API request
  response <- POST(
    url = "https://api.anthropic.com/v1/messages",
    headers,
    body = toJSON(request_body, auto_unbox = TRUE)
  )
  
  # Check if request was successful
  if (http_status(response)$category != "Success") {
    stop(paste("API request failed:", http_status(response)$message, 
               "\nDetails:", content(response, "text", encoding = "UTF-8")))
  }
  
  # Parse response and extract text content
  result <- fromJSON(content(response, "text", encoding = "UTF-8"))
  return(as.character(result$content)[2])
}

The only argument that needs to be supplied to the function is the prompt that you want to send to the model. This should be a text string that is enclosed by parentheses: “Tell me a joke about educational measurement.”

test_joke <- call_claude("Tell me a joke about educational measurement.")

test_joke

[1] "Why did the student bring a ladder to take the standardized test?\n\nBecause they heard the questions were on different *levels*!\n\n---\n\n(And if you want a more inside-baseball version: A psychometrician walks into a bar. The bartender asks, \"What'll you have?\" The psychometrician replies, \"I can't decide—but I can tell you with 95% confidence that my true preference falls within a two-drink interval.\")"

A few things to note:

1. Generally speaking, the process used by a generative AI model involves predicting the next token to be selected and sampling from a distribution of possible options. (We will discuss ways to control this later.) A (sometimes, mostly) beautiful implication of this is that responses from generative AI models will often be different, even if the exact same prompt is used. This is especially the case in the simple function above, as we haven’t made an effort to tune the model generation parameters to achieve a deterministic or replicable result. Thus, when you call Anthropic and have it generate a joke for you using the same prompt, you will likely get different responses each time. In fact, each time that I compile this book on GitHub the result changes!

2. You may notice that escape sequences (\n, \t, etc.) may be present because the information is returned in the JSON format. One way to handle this is to wrap your response in cat(). We’ll revisit this again later in other sections.

cat(test_joke)

Why did the student bring a ladder to take the standardized test?

Because they heard the questions were on different *levels*!

---

(And if you want a more inside-baseball version: A psychometrician walks into a bar. The bartender asks, "What'll you have?" The psychometrician replies, "I can't decide—but I can tell you with 95% confidence that my true preference falls within a two-drink interval.")