Skip to main content
NowFlowAgentic Application Layer
HomeDemoDocsBlog
Sign In

Docs

11 pages

Explore
Docs Home1
  • Overview
Getting Started2
  • Quick Start
  • Documentation Map
Platform5
  • Workflow System
  • Core Capabilities
  • Deployment
  • Resources & Knowledge
  • Observability
Operations2
  • Security & Compliance
  • Authentication
Blocks1
  • Blocks Catalog

Navigation

Mobile Docs
Final CallBuild with intent

Ready to Build with NowFlow?

Build once, publish everywhere with a unified runtime for enterprise and product teams.

Get Started Free
Agentic Workflow BuilderAI Agents300+ IntegrationsAPI DeployEmbedded UIChat SurfacesWorkflow VersioningLive CanvasHuman-in-the-LoopTeam WorkspacesEnterprise ControlsAudit LogsRuntime ObservabilityMulti-Surface LaunchAgentic Workflow BuilderAI Agents300+ IntegrationsAPI DeployEmbedded UIChat SurfacesWorkflow VersioningLive CanvasHuman-in-the-LoopTeam WorkspacesEnterprise ControlsAudit LogsRuntime ObservabilityMulti-Surface Launch
NowFlow

Agentic workflow platform by Neura Parse Ltd.

Product

  • Workflow Builder
  • AI Agents
  • Integrations
  • Pricing

Company

  • Company
  • Blog
  • Security
  • Demo Request
  • Contact

Legal

  • Legal
  • Privacy Policy
  • Terms of Service
  • Cookie Policy
  • DPA

© 2026 Neura Parse Ltd. All rights reserved.

Privacy·Terms·Cookies
Catalog

Agentic RAG

Advanced retrieval with iterative refinement and multi-source search

Agentic RAG goes beyond simple retrieval by iteratively refining queries, searching multiple sources, evaluating result quality, and re-ranking results. Supports dense, sparse, and hybrid retrieval strategies.

Agentsagentic_rag

Usage

  1. 01Add the block to your workflow and connect it to the upstream step.
  2. 02Configure any required credentials or tokens in the inputs.
  3. 03Fill in required inputs and optional parameters for the run.
  4. 04Run a test execution, inspect outputs, and iterate before deploying.
  5. 05Deploy the agentic_rag block with monitoring enabled in production.

Inputs (UI)

Query

long-input

Placeholder: Enter the question or search query...

Layout: full

Knowledge Source IDs

short-input

Placeholder: Comma-separated source IDs or use knowledge-source-input

Layout: full

Retrieval Strategy

dropdown

Layout: half

Options: Hybrid (Recommended), Dense (Embedding), Sparse (Keyword)

Max Results

slider

Layout: half

Enable Re-ranking

switch

Layout: half

Iterative Refinement

switch

Layout: half

Max Iterations

slider

Layout: half

Condition: iterativeRefinement = true

Inputs (API)

query

string

Required

sourceIds

string

Required

strategy

string

Optional

maxResults

number

Optional

reranking

boolean

Optional

iterativeRefinement

boolean

Optional

maxIterations

number

Optional

Outputs

Primary response type:

json

Tool Access

agentic_rag_search