UltraRAG

Active
GitHub Python Apache-2.0

Description

A low-code MCP framework for building complex and innovative RAG pipelines. Combines visual pipeline design with MCP protocol integration for end-to-end RAG — from data ingestion and chunking to retrieval and generation.

Key Features

  • First lightweight RAG framework built on MCP architecture
  • Low-code YAML configuration for complex RAG workflow orchestration
  • Core components (retriever, generator, etc.) as independent MCP Servers
  • Built-in unified evaluation system with mainstream research benchmarks
  • Visual Pipeline Builder with bidirectional sync between Canvas and code

Use Cases

💡 Rapid end-to-end RAG system setup: ingestion, chunking, retrieval to generation
💡 RAG research experiments: standardized evaluation workflows and benchmark comparison
💡 Interactive RAG prototyping: pipeline logic converts to conversational Web UI in one click
💡 Knowledge base Q&A: build custom knowledge bases for document Q&A

Strengths & Limitations

Strengths

  • Actively maintained, recent updates
  • High community interest (5.7k stars)
  • Permissive open-source license (Apache-2.0)
  • Responsive to issues, low backlog

Quick Start

git clone https://github.com/OpenBMB/UltraRAG.git --depth 1 && cd UltraRAG && pip install uv && uv sync

Related Projects

Unstract

7.2k · Python
Active A+

LLM-driven extraction of unstructured data, built for API deployments and ETL pipeline workflows. Automates document parsing, PDF extraction, and intelligent data processing with LLM-powered intelligence.

data-processingragpython +3
  • · Prompt Studio defines document extraction schemas with natural language
  • · Supports REST API deployment and ETL pipeline workflows
  • · MCP Server integration connecting to Claude and other AI agents

Instill Core

2.3k · Python
Stale C

A full-stack AI infrastructure tool for data, model, and pipeline orchestration. Streamlines building versatile AI-first applications with a visual pipeline editor for end-to-end workflows from data ingestion to model inference.

workflowpythonautomation +3
  • · End-to-end AI platform for data, pipeline, and model orchestration
  • · Visual pipeline editor for building versatile AI-first APIs and automated workflows
  • · Unstructured data processing turning documents, images, audio, and video into AI-ready formats

Ouroboros

5.8k · Python
Active A

Ouroboros is a spec-driven multi-agent framework that shifts from traditional prompting to specification-driven development, supporting multi-agent collaboration, MCP tool integration, and automated workflow orchestration for building high-quality agent systems.

multi-agentframeworkagent +4
  • · Spec-first workflow: Socratic interview → immutable seed spec → execution → evaluation
  • · Multi-runtime adapter supporting Claude Code, Codex CLI, OpenCode, Hermes, Gemini, Kiro, Copilot
  • · Three-layer architecture: Kernel (contract engine), Plugins (domain workflows), Shell (TUI cockpit)