What is OpenDocuments?

OpenDocuments is an open source, self-hosted RAG platform for AI document search. It connects to scattered documents, indexes them, and answers natural-language questions with source citations.

OpenDocuments is designed for teams that want a private knowledge base search system without sending every document to a hosted enterprise search vendor. It can run locally with Ollama or connect to cloud model providers such as OpenAI, Anthropic, Google, and xAI.

What problem does OpenDocuments solve?

Teams often store knowledge across many tools:

• GitHub repositories and wikis
• Notion pages and databases
• Google Drive files
• Confluence spaces
• S3 or Google Cloud Storage buckets
• Swagger and OpenAPI specs
• Local Markdown, PDF, DOCX, XLSX, CSV, PPTX, code, and email files
• Internal or public web pages

OpenDocuments gives those documents one AI-searchable layer. Users can ask questions like "How does authentication work?", "What changed in the v3 product spec?", or "What is our remote work policy?" and get grounded answers with source references.

How does OpenDocuments work?

OpenDocuments follows a production RAG flow:

1. Connect to document sources.
2. Parse files and preserve useful structure.
3. Chunk documents for retrieval.
4. Generate embeddings.
5. Store metadata in SQLite and vectors in LanceDB.
6. Search with hybrid vector and keyword retrieval.
7. Rerank and expand results based on the selected RAG profile.
8. Generate an answer with citations and confidence signals.

Who is OpenDocuments for?

OpenDocuments is useful for:

• Engineering teams that need AI search over code, API docs, runbooks, and architecture notes
• Product and operations teams that need answers from specs, spreadsheets, policies, and meeting notes
• AI-assisted development teams that want Claude Code, Cursor, Windsurf, or another MCP client to search internal knowledge
• Organizations that prefer self-hosted infrastructure and local model options
• Developers building custom RAG workflows in TypeScript

What does OpenDocuments include?

OpenDocuments includes:

• Web UI for chat, documents, connectors, plugins, settings, and admin views
• CLI for indexing, asking, searching, diagnostics, auth, backup, and automation
• HTTP API and TypeScript SDK
• MCP server for AI coding assistants
• Plugin system for parsers, connectors, model providers, and middleware
• Team mode with API keys, roles, rate limits, PII redaction, audit logs, security alerts, OAuth SSO, and workspace isolation

Short answer

OpenDocuments is a private AI search engine for your organization's documents. It is best described as a self-hosted RAG platform, an open source enterprise search alternative, and a knowledge base for AI coding assistants.