# Ragen AI

- [Introduction to Ragen AI: Self-Hosted RAG Platform](https://docs.ragen.ai/introduction.md): Ragen AI turns your company documents into an AI knowledge base with multilingual retrieval, source-traceable answers, and an OpenAI-compatible API.
- [Get Started with Ragen AI: Up and Running in Minutes](https://docs.ragen.ai/quickstart.md): Scaffold a local Ragen AI instance with create-ragen-app, start your services, and ask your first question of your own documents in under ten minutes.
- [Self-Hosting Ragen AI on Your Own Infrastructure Guide](https://docs.ragen.ai/self-hosting.md): Deploy Ragen AI with Docker Compose on your own servers — covering services, environment variables, feature flags, and production hardening.
- [Core Concepts: Projects, Organizations, and API Keys](https://docs.ragen.ai/concepts/overview.md): Understand projects, organizations, threads, and API keys — the core building blocks that shape every interaction you have with Ragen AI.
- [Knowledge Base: Folders, Versioning, and File Visibility](https://docs.ragen.ai/concepts/knowledge-base.md): Organize documents into nested folders with fine-grained permissions. Control who sees each file at the team, user, and org level.
- [RAG Pipeline: How Ragen AI Retrieves and Generates Answers](https://docs.ragen.ai/concepts/rag-pipeline.md): Learn the four-stage pipeline — rephrase, hybrid search, optional rerank, generate — that turns your documents into accurate, cited answers.
- [Ragen AI Access Control: Roles, Keys, and Permissions](https://docs.ragen.ai/concepts/access-control.md): Understand org roles, platform roles, document permissions, and how Ragen AI enforces access at retrieval time — not just in the UI.
- [Complete Environment Variable Reference for Ragen AI](https://docs.ragen.ai/configuration/environment.md): Configure every aspect of Ragen AI through environment variables — storage, encryption, mail, models, auth, and feature flags, all in one place.
- [Configuration reference](https://docs.ragen.ai/configuration/environment-reference.md): Every environment variable Ragen reads, generated from the schemas the applications validate against at boot.
- [Configuring AI Language Model Providers in Ragen AI](https://docs.ragen.ai/configuration/models.md): Set which model answers, rephrases, summarises and embeds — and reach any provider directly or through a gateway you run.
- [Model Gateway — How Ragen Reaches Model Providers](https://docs.ragen.ai/configuration/model-gateway.md): Ragen calls model providers directly through a route table. Attach Portkey, vLLM, Ollama or any OpenAI-compatible endpoint per model.
- [Configuring Local and S3-Compatible Storage in Ragen AI](https://docs.ragen.ai/configuration/storage.md): Store uploaded documents on the local filesystem by default, or switch to any S3-compatible provider for multi-replica and production deployments.
- [Run Ragen AI Fully Offline with Local LLM Model Servers](https://docs.ragen.ai/configuration/open-models.md): Point all four model roles at a local vLLM or Ollama server so no document content, query text, or embeddings ever leave your network.
- [Connect External Services with Ragen MCP Integrations](https://docs.ragen.ai/integrations/mcp-connectors.md): Connect Google Drive, Slack, HubSpot, Gmail, ClickUp, and more as live data sources inside Ragen conversations using MCP connectors.
- [Use Your Ragen AI Assistants from Any MCP-Compatible Client](https://docs.ragen.ai/integrations/mcp-server.md): Connect Claude Desktop, Cursor, or a custom agent to your Ragen assistants using the MCP Server — no REST API integration required.
- [Configure OAuth Sign-In and Authentication in Ragen AI](https://docs.ragen.ai/integrations/oauth.md): Configure email, magic-link, and Google OAuth sign-in for the Ragen main app and admin panel, including allowed domain restrictions.
- [Security and Data Privacy in Ragen AI: Full Overview](https://docs.ragen.ai/security/overview.md): Understand where your data lives, what leaves your network, and how Ragen enforces access control, audit logging, and prompt safety.
- [Encrypting Conversation Thread Data at Rest in Ragen](https://docs.ragen.ai/security/encryption.md): Enable AES-256-GCM envelope encryption for conversation threads at rest using Scaleway Key Manager, AWS KMS, or a local master key.
- [Masking Personal Data Before It Reaches the Model](https://docs.ragen.ai/security/pii-masking.md): Detect and mask personal data with Microsoft Presidio before it reaches an LLM or the vector store, with Polish recognisers that validate checksums.
- [Managing Your Ragen Installation with the Admin Panel](https://docs.ragen.ai/security/admin-panel.md): Use the Ragen admin panel to manage organisations, users, API keys, feature flags, usage, and security events across your installation.
- [Ragen AI Roadmap: Current and Planned Upcoming Features](https://docs.ragen.ai/roadmap.md): Explore what Ragen AI is actively building right now, what is planned for the next milestone, what comes later, and what is out of scope.
- [Frequently Asked Questions About Ragen AI Setup and Use](https://docs.ragen.ai/faq.md): Answers to common questions about setting up Ragen AI, choosing AI models, keeping your data private, and building API integrations.
- [Ragen AI API Quickstart — SDK Setup and First Call](https://docs.ragen.ai/api-reference/quickstart.md): Install the TypeScript SDK, send your first chat completion, and stream tokens from your self-hosted Ragen AI instance in under five minutes.
- [Ragen AI API Authentication — API Keys and Headers](https://docs.ragen.ai/api-reference/authentication.md): Set up API key authentication for Ragen AI. Learn the Bearer token format, Authorization header usage, and how to handle 401 and 403 errors.
- [Chat API — Single-Turn RAG Question and Answer Endpoint](https://docs.ragen.ai/api-reference/chat.md): POST /v1/chat — Send a single message to a Ragen assistant and get a RAG-grounded answer back. Supports streaming via Server-Sent Events.
- [Chat Completions — OpenAI-Compatible Ragen RAG API](https://docs.ragen.ai/api-reference/chat-completions.md): POST /v1/chat/completions — OpenAI-compatible endpoint that runs the full Ragen RAG pipeline with multi-turn conversation support.
- [Files API — Upload and Manage Knowledge Base Documents](https://docs.ragen.ai/api-reference/files.md): POST, GET, DELETE /v1/files — Upload documents to a Ragen project, poll for processing status, and manage your knowledge base files.
- [Assistants API — Create and Manage Ragen AI Projects](https://docs.ragen.ai/api-reference/assistants.md): CRUD endpoints for Ragen assistants (projects). OpenAI-compatible API for listing, creating, updating, and deleting knowledge base projects.
- [Threads and Messages API — Conversation Persistence](https://docs.ragen.ai/api-reference/threads.md): OpenAI-compatible endpoints for creating threads and storing messages. Does not run the model — use Chat Completions for AI generation.
- [Search API — Query the Ragen AI Knowledge Base Directly](https://docs.ragen.ai/api-reference/search.md): POST /v1/search — Retrieve ranked document chunks from your Ragen knowledge base without generating an AI answer. Returns context and source file IDs.
- [Ragen AI TypeScript SDK — Methods, Config & Examples](https://docs.ragen.ai/api-reference/typescript-sdk.md): Complete reference for @webamigos/ragen-sdk-ts: chat completions, streaming, file upload, assistants, threads, and TypeScript error handling.
- [Ragen MCP Server: Connect Claude Desktop and AI Agents](https://docs.ragen.ai/api-reference/mcp-server.md): Use the Ragen MCP Server to connect Claude Desktop, Cursor, or any AI agent to your Ragen assistants and knowledge base over Streamable HTTP.