AgenticAIAgentAgenticRag icon

AgenticAIAgentAgenticRag

Comprehensive Agentic RAG System by BigBoss

Actions2

Agent → Ask

AI-generated

Summary

Ask the agent a question using RAG, performing iterative research with optional web search, vector store, and tool usage.

Inputs

  • Main — Input items to process; each item's index is used for parameter evaluation.
  • Model (required) — Language model used for planning, reflection, and answer synthesis.
  • Vector Store — Optional vector store for retrieving knowledge when RAG source includes vector store.

Output shape

array of objects per input item

Each output item contains the fields below under its json property.

Examples

Example 1: Basic question answering with default settings

Query: 'What is the capital of France?', System Prompt: default, Max Iterations: 5, Confidence Threshold: 0.85, RAG Source: Vector Store, Search Strategy: Adaptive, Enable Memory: true, Tools: []

Links

Discussion