Source: Notebook experiment 22: Agentic RAG
Runnable scaffold where retrieval is a tool a ReAct agent calls on demand.
Contents
- What: runnable experiment in the owner’s public Notebook repo
- Author: Loic (coolloic)
- Date: undated (read 2026-11-14)
- Original: experiments/22-agentic-rag
Summary
A LangGraph create_react_agent with a search_policy retriever tool over FAISS. The agent decides whether and what to retrieve, in contrast to experiment 13 which always retrieves.
Key claims
- Agentic RAG turns retrieval into a tool rather than a fixed pipeline step.
- It suits mixed workloads where some questions need the knowledge base and others need several targeted lookups.