Introduction

Egret is a RAG-powered compliance advisory platform that retrieves relevant regulatory passages and generates cited answers. Built for financial services, healthcare, legal, and other regulated industries.

What is Egret?

Egret combines retrieval-augmented generation (RAG) with source citations to help compliance teams navigate complex policy landscapes. Instead of keyword search or uncited AI responses, Egret:

• Retrieves the most relevant passages from your regulatory document corpus using vector similarity and keyword matching
• Generates natural-language answers with inline source citations
• Guarantees zero-training-exposure — your data is never used to train or fine-tune any model

Key features

• Knowledge Domains — Scoped document collections for different regulatory areas (HIPAA, SOC 2, GDPR, etc.)
• Streaming responses — Real-time token delivery via Server-Sent Events
• Source citations — Every claim links back to its source document and paragraph
• Multi-tenant organizations — Role-based access with Owner, Admin, and Member roles
• REST API — Full API with typed TypeScript client (@egret/api)
• AES-256 encryption — All stored data encrypted at rest

How it works

1. Ingest — Upload regulatory documents or connect knowledge domains. Egret chunks, embeds, and indexes everything.
2. Query — Ask a natural-language question scoped to one or more domains.
3. Generate — The LLM synthesises a response with source citations from retrieved passages.

Next steps

Ready to get started? Follow the Quickstart guide to make your first query in under 5 minutes.