Agentic RAG

The agent decides autonomously whether, when, and how to retrieve — issuing and reformulating queries, selecting among sources, grading the relevance of returned passages, and iterating or skipping retrieval — instead… Typical conditions for use: Some queries need external knowledge and others do not; A single retrieval pass returns noisy or insufficient con…

Use this profile to understand the building block briefly, place it in the model, and switch to the 360° assessment when needed.

Content type

Concept

Theoretical construct: explains a term, principle, or mental model.

Classification level

Structure

What organizes, connects, or makes decisions possible.

Available perspectives

Open this building block directly in an already curated view.

360°

In brief

Agentic RAG in brief

Agentic RAG combines retrieval-augmented generation with an agent that plans searches, selects sources, and iteratively improves retrieval.

Where does it come from?

Agentic RAG grew from combining classic RAG pipelines with agentic planning; research on modular multi-agent RAG describes specialized agents for querying, selection, extraction, and checking.

The mental model

A research team frames a question, finds documents, checks evidence, and follows up when the first search leaves gaps.

Key building blocks

Core idea

The agent steers retrieval and uses its evidence in the answer.

Use

Iterative search helps with complex questions that have incomplete context.

Practice context

Querying, document selection, evidence extraction, and quality checks form a controlled flow.

When is it useful?

Agentic RAG improves traceability and retrieval quality when one vector search is insufficient.

Multi-Agent Systems: Agentic RAG MAO-ARAG paper

Why is this building block relevant?

The agent decides autonomously whether, when, and how to retrieve — issuing and reformulating queries, selecting among sources, grading the relevance of returned passages, and iterating or skipping retrieval — instead…