Skip to content
Forward Deployed
Forward Deployed
Agentic RAG: sections and modules
T9.4

RAG as an Agent Tool

Expose a RAG pipeline as a tool the agent chooses to call in LangGraph.

Lessons
5 lessons
By
Sourav Banerjee
Prerequisites
None

Lessons

Start module
5 lessons
  1. 1
    RAG as an Agent Tool: overview

  2. 2
    RAG as a tool inside a LangGraph agent

  3. 3
    Managed RAG platforms: Contextual AI

  4. 4
    Exercise: LlamaIndex Wikipedia agent with Chainlit

  5. 5
    RAG as an Agent Tool: interview questions

Practise and Check

Deliverable
Two-tool agent: give an agent a private-docs RAG tool and a web-search tool, and tune the tool descriptions until it picks the right one on 20 mixed questions.
Assessment
Answer the five Interview questions without notes, then run the lab.
Interview probe
Larkspur adds a `web_search` tool to the support agent. A week later a customer asks "Can I get a refund on my annual plan?" and the agent quotes another provider's refund policy. How do you find out what happened?
Common mistakes
  • Tool descriptions that never say where each tool's scope stops
  • Trusting tuned tool descriptions without checking held-out questions
  • Keeping the Chainlit index and agent in module globals

Your Progress

How confident are you with this module?

1 means "I could not explain it yet"; 5 means "I could teach it". Rate again any time.

 

Saved in this browser when you leave the box.

Related Modules