RAGFlow

An agent-native data platform for enterprise knowledge. Build knowledge bases from your own documents and serve them to AI agents with citations.

This page describes the application, why it requests access to your Google Account, and how you stay in control of your data. No account or sign-in is required to read it.

Purpose of the application

RAGFlow is used by teams to import their own business documents, build enterprise knowledge bases with hybrid retrieval, and run AI agents that answer questions from that content with traceable citations. Typical uses are enterprise search, document question answering, knowledge management and agent automation.

RAGFlow is not a consumer social network, and it is not an image, audio or video generation product.

What RAGFlow is

RAGFlow turns scattered enterprise content into a single queryable context layer. Documents are parsed, chunked and indexed, and retrieval combines vector search, full-text search and re-ranking so that answers can cite the exact source passage.

The platform is agent-native: the same knowledge base is exposed to agents through APIs and tools, so an agent harness — the orchestration layer that plans, calls tools and produces answers — can retrieve grounded context instead of guessing. RAGFlow ships with its own agent harness and also works with external harnesses you already run.

What you can do with RAGFlow

Enterprise knowledge bases

Upload or connect your own documents, images and tables, then turn them into traceable knowledge with hybrid retrieval and cited answers.

Agent-native by design

The knowledge base is exposed to agents through APIs and tools, so reasoning workflows retrieve grounded context instead of guessing.

Built-in agent harness

Compose retrieval, tools, models and MCP servers into agents with visual workflows and run them where your data lives.

Open and self-hostable

Use the hosted service, or deploy the open-source edition in your own environment when data residency or isolation is required.

How RAGFlow uses Google user data

Google integrations are optional and activate only after you authorize them. RAGFlow requests only the scopes required for the features you turn on.

Sign in with Google

Data accessed: Name, email address and profile picture
Scopes: openid, email, profile
Why: Create, identify and secure your RAGFlow account, and keep your workspace access under your control.

Google Drive import

Data accessed: Only the files and folders you explicitly select, plus their metadata and content
Scopes: Google Drive read-only
Why: Import that content into a knowledge base, index it, and keep it in sync so you can search and ask questions about it.

What we never do with Google user data:

We do not use it for advertising or ad personalization.
We do not sell it.
We do not use it to train or improve generalized AI or machine learning models.
We do not transfer it to third parties except as needed to provide the feature you requested, to comply with law, or as part of a merger or acquisition.
We do not allow humans to read it, except with your explicit consent, for security purposes, or to comply with law.

RAGFlow's use of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements.

Prohibited uses

RAGFlow prohibits the use of the Services to create, request, store or distribute AI-generated non-consensual intimate imagery (NCII), child sexual abuse material (CSAM), or any other sexual content involving minors or real people without their consent.

We do not use Google APIs to generate, host or distribute such content, and we suspend accounts that violate this policy. Our Terms of Use set out these restrictions and their consequences.

Your control over your data

Revoke RAGFlow's access to your Google Account at myaccount.google.com/permissions.
Disconnect an integration in your RAGFlow workspace at any time; synchronization stops immediately.
Request deletion of imported content and stored tokens by emailing privacy@ragflow.io. We delete the associated data within 30 days, except where we are legally required to retain it.

Full details are in our Privacy Policy.

Operator and policies

RAGFlow is operated by InfiniFlow Inc. Questions about this page, the application, or our data practices can be sent to privacy@ragflow.io or through our contact page.

Registered address: Palm Grove Unit 4, 265 Smith Road, George Town, P.O. Box 52A Edgewater Way, #1653, Grand Cayman KY1-9006, Cayman Islands.

Last updated: 2026-10-20