What problem it solves

Search tools commonly return file paths and matching lines, leaving users to open files and synthesize the context themselves. AutoRAG Agent searches files, mail, chats, and notes in place, verifies answers against source documents, and can also organize or edit files on request.

Key features

• Federated search across local sources — The agent searches files, mail, chats, and notes where the data already lives rather than requiring it to be migrated into a centralized index. Configured external sources include Slack, KakaoTalk, Discord, Telegram, WhatsApp, Mail, GitHub, and Obsidian.

  Source: README

• Grounded answers with citations — A single configured model searches the available sources, opens promising files directly to verify facts, and curates the result into numbered knowledge units with citations.

  Source: README

• Multiple retrieval methods — The engine supports learned-sparse keyword search, semantic vector search, hybrid ranking, local file-name discovery, and federated datasource retrieval.

  Source: README

• Interactive and programmatic access — The librarian can be used through a terminal interface, messenger connections, a CLI, an MCP server, and a TypeScript API. This lets users choose interactive search sessions, scripting, agent integration, or application embedding.

  Source: README, src

• Local-first semantic indexing — Semantic indexing runs through a locally provisioned embedding engine, while file contents are not sent to a remote embedder unless that is explicitly configured. Sign-in tokens and API keys are kept in the system's encrypted store.

  Source: README

Quick start

Copied verbatim from the project's README.

# Install CLI globally
bun install -g @autorag/librarian
# or with npm:
npm install -g @autorag/librarian

# Add as a TypeScript/JavaScript library
bun add @autorag/librarian

# Initialize configuration for your documents folder
autorag init --search-paths ~/Documents/research

# Korean + English is the default; set it explicitly for other corpora
autorag init --search-paths ~/Documents/research --languages ja,en

# Index documents (parses PDFs/Markdown, builds MinSync sparse and dense vectors)
autorag refresh

# Check indexing health
autorag status

Who it may suit

This may suit office workers who need to search and act on information spread across local documents and communication sources. It may also suit developers who want retrieval available through a CLI, TypeScript API, MCP server, or coding-agent skills.

Topics to discuss

• Which office workflows would benefit most from a librarian that can also organize and edit files?

• How should retrieval memory be used across changing projects without letting prior context override current evidence?

• Which datasource and messaging integrations are the most important to add or refine next?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/Marker-Inc-Korea/AutoRAG#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.