Code Generation Agent • LLMAgentR

What This Agent Does

build_code_agent() helps you generate or debug R code from natural-language instructions.

Workflow Diagram

Generate Mermaid PNGs

library(LLMAgentR)

my_llm_wrapper <- function(prompt, verbose = FALSE) "LLM response placeholder"

workflow <- build_code_agent(
  llm = my_llm_wrapper,
  output = "both",
  direction = "LR"
)

save_mermaid_png(
  x = workflow,
  file = "pkgdown/assets/code-agent-workflow.png"
)

Step 1: Create an LLM Wrapper

my_llm_wrapper <- function(prompt, verbose = FALSE) {
  # Replace with your provider call (OpenAI, Groq, Anthropic, etc.)
  "LLM response placeholder"
}

Step 2: Build a Reusable Agent

library(LLMAgentR)

coder <- build_code_agent(
  llm = my_llm_wrapper,
  max_tries = 3,
  backoff = 2,
  verbose = FALSE
)

Step 3: Run a Task

result <- coder(
  "Write an R function that standardizes all numeric columns in a data frame."
)

str(result)

Notes for Beginners

The returned object is a normal R list.
Main fields to inspect: success, attempts, llm_response.
You can reuse coder for many prompts.