--- title: Make Your Own LLM Agent in R | daily.dev description: You can build your own LLM agent in R using the ellmer package to connect to a model. The approach: define tools as R functions the agent can call, write a... canonical: https://daily.dev/posts/make-your-own-llm-agent-in-r-ve3mdlusa twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Make Your Own LLM Agent in R | daily.dev og:description: You can build your own LLM agent in R using the ellmer package to connect to a model. The approach: define tools as R functions the agent can call, write a... og:url: https://daily.dev/posts/make-your-own-llm-agent-in-r-ve3mdlusa og:image: https://api.daily.dev/og/posts/vE3MDlUSA.png og:image:alt: Make Your Own LLM Agent in R og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Make Your Own LLM Agent in R **[R-bloggers](https://daily.dev/sources/rbloggers)** · 3 min read · 0 upvotes · 0 comments ## Summary You can build your own LLM agent in R using the ellmer package to connect to a model. The approach: define tools as R functions the agent can call, write a parser that detects tool use in the LLM's response, then run it all in a while loop so the agent consumes its own output. To demonstrate registered tools, the author built a package called turbochook with three functions: handfish() for manual, step-by-step approval, turbochook() for a fully autonomous agentic loop, and ralph() for a self-completing todo list loop. It's alpha software with no cybersecurity guarantee, so run it in a sandbox. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do I build a simple LLM agent in R? Use the ellmer package to connect to an LLM, then define tools as R functions the agent can call, such as writing and running R code. Write a parser that checks the LLM's response for tool use and executes matching tools, then wrap this in a while loop so the agent consumes its own output repeatedly. Registering tools formally with ellmer reduces syntax mistakes when the agent invokes them. _Building an agent loop in R is easier to reason about with daily.dev's coverage of ellmer and agent tooling nearby._ ### What is the turbochook R package and what does it do? turbochook is an alpha-stage R package providing three agent functions: handfish(), which requires manual approval at every step for beginners and the security conscious; turbochook(), which runs an agentic loop fully autonomously without asking permission; and ralph(), which runs a self-completing todo list loop. It was built with Claude Code and installs from GitHub, with no cybersecurity guarantee, so it should run in a sandbox or isolated container. _Developers evaluating experimental R agent packages can track releases like this one on daily.dev._ ## Similar posts on daily.dev - [How to keep enjoying programming in a world of LLMs](https://daily.dev/posts/how-to-keep-enjoying-programming-in-a-world-of-llms-dbajtmasp) · Lobsters · 0 upvotes · 0 comments - [From Local LLM to Tool-Using Agent](https://daily.dev/posts/from-local-llm-to-tool-using-agent-kmmln7rio) · Towards Data Science · 1 upvotes · 0 comments - [How I write software with LLMs](https://daily.dev/posts/how-i-write-software-with-llms-jekylkhup) · Hacker News · 2 upvotes · 0 comments - [Automating Web Research](https://daily.dev/posts/automating-web-research-pwuhcbqf7) · LangChain · 0 upvotes · 0 comments - [RubyLLM 1.12: Agents Are Just LLMs with Tools](https://daily.dev/posts/rubyllm-1-12-agents-are-just-llms-with-tools-scy0gxemk) · RUBYLAND · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#r](https://daily.dev/tags/r), [#claude-code](https://daily.dev/tags/claude-code) [View this post on daily.dev](https://daily.dev/posts/make-your-own-llm-agent-in-r-ve3mdlusa) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Make Your Own LLM Agent in R","url":"https://daily.dev/posts/make-your-own-llm-agent-in-r-ve3mdlusa","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/make-your-own-llm-agent-in-r-ve3mdlusa"},"datePublished":"2026-10-25T22:47:26.166Z","dateModified":"2026-10-25T22:47:54.974Z","description":"You can build your own LLM agent in R using the ellmer package to connect to a model. The approach: define tools as R functions the agent can call, write a...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/499220ca6df7394f27f01806792a5f8b?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/499220ca6df7394f27f01806792a5f8b?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"R-bloggers","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"R-bloggers","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/98bcbbdf72774292a55ceda8f2277f1e","url":"https://daily.dev/sources/rbloggers"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/make-your-own-llm-agent-in-r-ve3mdlusa","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,r,claude-code","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"R-bloggers","item":"https://daily.dev/sources/rbloggers"},{"@type":"ListItem","position":3,"name":"Make Your Own LLM Agent in R"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/make-your-own-llm-agent-in-r-ve3mdlusa#faq","mainEntity":[{"@type":"Question","name":"How do I build a simple LLM agent in R?","acceptedAnswer":{"@type":"Answer","text":"Use the ellmer package to connect to an LLM, then define tools as R functions the agent can call, such as writing and running R code. Write a parser that checks the LLM's response for tool use and executes matching tools, then wrap this in a while loop so the agent consumes its own output repeatedly. Registering tools formally with ellmer reduces syntax mistakes when the agent invokes them. Building an agent loop in R is easier to reason about with daily.dev's coverage of ellmer and agent tooling nearby."}},{"@type":"Question","name":"What is the turbochook R package and what does it do?","acceptedAnswer":{"@type":"Answer","text":"turbochook is an alpha-stage R package providing three agent functions: handfish(), which requires manual approval at every step for beginners and the security conscious; turbochook(), which runs an agentic loop fully autonomously without asking permission; and ralph(), which runs a self-completing todo list loop. It was built with Claude Code and installs from GitHub, with no cybersecurity guarantee, so it should run in a sandbox or isolated container. Developers evaluating experimental R agent packages can track releases like this one on daily.dev."}}]} ```