## Clarification Can you clarify what do you mean by: > You would think someone would make a really great library to create AI Agents using Raku. Maybe, you

## Clarification Can you clarify what do you mean by: > You would think someone would make a really great library to create AI Agents using Raku. Maybe, you mean using Raku to make: 1. Standalone, web based LLM agents 2. Plug-ins for ChatGPT (or other LLM providers) 3. LLM Retrieval Augmented Generation (RAG) functionalities, or services, or objects LLM RAG (point 3) is supported in Raku, see: "Raku RAG demo", https://youtu.be/JHO2Wk1b-Og https://youtu.be/JHO2Wk1b-Og . Points 1 and 2 should be Raku-addressed sooner or later, but Raku provides nice alternatives to them. (See below.) ## LLMs at your fingertips I would say, with the Raku LLM packages one can create "AI agents" on the spot: - Programmatic ways to configure and execute LLM functions and chat objects. - Using easy to compose LLM functions and easy to access prompts. - Via Jupyter based notebooks with LLM-ready magic cells, aka "chatbooks." In one chatbook multiple, global chats can be held with multiple LLM personas. Also, each of the LLM-service access packages provides a Command Line Interface (CLI) script. That is another quick, "on the spot" type of access to AI. ## References Please see these Raku LLM packages: https://raku.land/?q=LLM https://raku.land/?q=LLM