rag-anythink-mcp

概述

RAG-Anything MCP Server 是一个生产就绪的 MCP 实现，结合了知识图谱查询、文档处理和多模态 AI 功能。它提供多种查询模式（朴素、局部、全局、混合、混合、绕过）以实现灵活的信息检索，能自动从文档中提取实体和关系，并使用 Neo4j 和带 pgvector 的 PostgreSQL 进行混合存储。该服务器支持处理 PDF 中的图像、表格和公式，适用于复杂的文档分析任务。

试试问 AI

装完之后，这里有 3 个你可以让 AI 做的事：

• ingest_document

  Ingest text or PDF documents

  试试:Add this research paper to the knowledge base试试:Process this PDF document for information extraction
• query_knowledge_graph

  Query with multiple modes (naive, local, global, hybrid)

  试试:Find information about machine learning using hybrid search试试:Search for topics related to renewable energy using global community search
• query_multimodal

  Query with images, tables, equations

  试试:Extract insights from the table in this PDF试试:Analyze the mathematical equation in this document
• process_document_file

  Process PDF files with multimodal extraction

  试试:Process this PDF file for text, images and tables试试:Extract all content from this research document
• insert_content_list

  Insert pre-parsed content

  试试:Add this pre-parsed document to the knowledge graph试试:Insert this content list into the database
• delete_data

  Delete documents by ID

  试试:Remove document with ID 12345 from the knowledge base试试:Delete this outdated research paper
• get_graph_statistics

  Get graph statistics

  试试:Show me statistics about the knowledge graph试试:How many nodes and relationships are in the graph?
• get_config_info

  Get configuration info

  试试:Show me the current configuration试试:What's the default query mode set to?

可对比工具

安装

# 克隆仓库
git clone https://github.com/serkanyasr/rag-anythink-mcp.git
cd rag-anythink-mcp

# 复制环境模板
cp .env.example .env

# 编辑 .env 文件并设置您的 OpenAI API 密钥
# OPENAI_API_KEY=sk-...

# 启动所有服务
docker-compose up -d

# 查看日志
docker-compose logs -f rag-mcp

{
  "mcpServers": {
    "rag-anything": {
      "command": "docker-compose",
      "args": ["up", "rag-mcp"],
      "env": {
        "OPENAI_API_KEY": "您的-api-密钥"
      }
    }
  }
}

rag-anythink-mcp 对比