repowisehkuds/rag-anything

hkuds / rag-anything

"RAG-Anything: All-in-One RAG Framework"

Python★ 23,485
Use in your agentAsk this codebase

github.com/hkuds/rag-anythingIndexed at a8538efUp to date with upstream

What is rag-anything?

rag-anything is an all-in-one multimodal document processing and retrieval system for documents containing text, images, tables, equations, charts, and multimedia. It parses mixed content, analyzes each modality, builds a multimodal knowledge graph, and supports unified querying. rag-anything is written in Python and built on LightRAG.

Written by a model from the README and the source

59 files and 481 symbols in 6 modules, written in Python.

Start reading

Code health

4.7out of 10Needs work

Hotspots average 2.2, the place fixes pay off most.

Full health report

Use hkuds/rag-anything in your agent

Point Claude Code, Cursor, Codex or VS Code at this repository. Your agent gets the index this page is built from: cited answers, callers, history and health for any file.

  1. 1.In Claude, open Customize → Connectors, click +, then Add custom connector. On a Team or Enterprise plan an owner adds it first under Organization settings → Connectors.
  2. 2.Paste the server URL below as the Remote MCP server URL, then click Add.
  3. 3.Click Connect. A Repowise sign-in window opens. Approve access.
  4. 4.Repowise's tools are now available in your Claude chats.

Connect with one click. Sign in with Repowise (OAuth), no API key to copy. Works in Claude on the web, desktop, and mobile.

Docs
bash
https://api.repowise.dev/mcp/hkuds/rag-anything
Editors need an API key: sign in to create one.Sign in to connect

This repo is public, so anyone with a Repowise account and an API key can connect.

Try asking

  • get_overview()
  • get_answer("how is rag-anything put together?")
  • get_why("why is rag-anything structured this way?")

Explore rag-anything

Key features of rag-anything

  • rag-anything provides an end-to-end multimodal pipeline from document ingestion and parsing to multimodal query answering.
  • rag-anything processes PDFs, Office documents, images, and diverse file formats.
  • rag-anything provides specialized processors for images, tables, mathematical equations, and heterogeneous content types.
  • rag-anything builds a multimodal knowledge graph with entity extraction and cross-modal relationship discovery.
  • rag-anything supports MinerU-based parsing, direct multimodal content injection, and direct insertion of pre-parsed content lists.
  • rag-anything provides hybrid retrieval across textual and multimodal content.

How hkuds/rag-anything works

Documents enter through the parsing workflow, which uses MinerU integration and format-specific handling to decompose content while preserving hierarchy and relationships. raganything/parser.py handles document parsing and content extraction. raganything/processor.py coordinates content categorization and concurrent text and multimodal processing. raganything/modalprocessors.py provides modality-aware processing for images, tables, mathematical expressions, and other content types. raganything/raganything.py provides the integrated RAG system, while raganything/query.py handles multimodal retrieval and querying. The resulting representations support knowledge-graph construction, cross-modal relationships, and query answers.

The codebase has 59 files and 481 symbols in 6 modules, led by Python. Code health is 4.7 out of 10, rated needs work.

Getting started with rag-anything

  1. Install the basic package from PyPI.
    pip install raganything
  2. Install all optional features for extended format support.
    pip install 'raganything[all]'

Commands copied from the README

hkuds/rag-anything FAQ

What is rag-anything?

rag-anything is an all-in-one multimodal document processing and retrieval system. It handles mixed documents containing text, images, tables, equations, charts, and multimedia.

How does rag-anything work?

rag-anything parses documents, routes content through modality-specific processors, builds a multimodal knowledge graph, and retrieves information across textual and multimodal content.

How do I install rag-anything?

rag-anything installs from PyPI with pip install raganything. Optional dependencies are available through extras such as raganything[all], raganything[image], and raganything[text].

What language is rag-anything written in?

rag-anything is written in Python. The system is built on LightRAG.

How is the rag-anything code organized?

rag-anything organizes its core code in raganything, with examples in examples, documentation in docs, and tests in tests. Core files include parser.py, processor.py, modalprocessors.py, raganything.py, and query.py.

Who is rag-anything for?

rag-anything is for academic research, technical documentation, financial reports, and enterprise knowledge management involving rich mixed-content documents.

What license does rag-anything use?

rag-anything uses the MIT license.

The summary, features and answers on this page are written by a model from the README and the source, and refreshed with each index.

About this page

Get a page like this for your own codebase

Repowise built this map of hkuds/rag-anything from its source and git history. It parses every symbol, draws the dependency graph, scores each file's health from complexity, duplication, coverage and churn, finds hotspots and owners in the history, and writes documentation for every module. Your coding agent reads the same index over MCP, so its answers cite real files.

Index your repo freeSee how Repowise worksFree for 2 public repos, with a page like this one.

Prefer local? pip install repowise && repowise init indexes your own checkout, no account needed.