Oboyu (覚ゆ)

MCP文档知识

提供方 sonesuke

文档知识 未标注 会话登录 社区 / 第三方 Claude DesktopClaudeCursorCodex
查看接入方式

Self-hosted MCP Japanese text indexing & search—chunking+embeddings with BM25×vector rerank

一句话选型结论

按资料字段整理

Oboyu (覚ゆ) 属于文档知识类 MCP 服务器,传输方式未标注,会话登录,可直接接入 Claude Desktop、Claude、Cursor 等,有可直接执行的安装命令。接入前先确认权限范围;涉及写入、删除或付款的操作,建议先在测试环境验证。

分类文档知识 传输方式未标注 鉴权会话登录 部署方式未标注
GitHub 星标 未标注 仓库热度,无公开仓库时显示未标注
工具数 未标注 暴露给模型的工具数量
资源 / 提示词 0 / 0 资源与提示词模板数量
维护方 社区 / 第三方 核验等级 仅资料核对

工具与能力

文档知识
数量来自来源目录的标注,具体工具清单以官方文档为准
工具数量未标注暴露给模型可调用的工具个数
资源数量0可读取的资源条目
提示词数量0内置提示词模板数量

能力标签

python search engine information retrieval japanese embeddings semantic search bm25 reranking

支持客户端

适用场景

pythonsearch engineinformation retrieval

安装与配置

未标注 会话登录
上游未给出传输方式
pip install oboyu

这条命令只是入口,具体启动参数请按官方文档补齐,不要直接照搬到别的服务器。

  1. 1
    准备运行环境

    按官方文档确认运行环境(常见为 Node.js 或 Python),装好后再执行安装命令。

  2. 2
    安装服务

    复制上面的安装命令执行;命令里带占位符的位置,换成自己的密钥或路径。

  3. 3
    写入客户端配置

    在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,让新配置生效。

  4. 4
    验证是否可用

    连接成功后随便调一个只读工具,确认能返回结果,再接入正式流程。

典型使用场景

3 类
按分类与标签整理的常见用法,具体以官方文档为准
文档检索问答

在你的文档库里按问题找答案。

知识库读写

把新资料写进知识库,随用随取。

内容整理

把零散资料归纳成结构化笔记。

选择建议

选之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 有现成的安装命令,不用自己找入口,照着跑就能起服务。
  • 已适配 Claude Desktop、Claude、Cursor,换客户端不用重新研究配置。

先想清楚的地方

  • 使用前要准备好会话登录对应的凭据,并按最小权限配置。
  • 资料没标注工具数量,实际能力要装好后确认。
  • 版本更新与行为细节以官方仓库为准。

更适合

  • 需要在文档知识场景里让模型直接调用外部能力的人。
  • 已经在用 Claude Desktop、Claude 的用户,接入路径最短。
  • 想先快速试一下、能接受在本地或自有环境跑服务的人。

不太适合

  • 不方便管理密钥或第三方授权的人。
  • 需要厂商级 SLA 和技术支持的生产环境。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
MCP 服务器工具数传输方式部署方式安装命令最后核验
Oboyu (覚ゆ)当前 未标注 未标注 未标注 有 未标注
dify 未标注 stdio · 本地进程 本地部署 有 2026-04-11
open-webui 未标注 stdio · 本地进程 本地部署 有 2026-04-11
JeecgBoot 未标注 stdio · 本地进程 本地部署 有 2026-04-11
Cherry Studio 未标注 未标注 未标注 无 2026-05-17
mempalace 未标注 stdio · 本地进程 本地部署 有 2026-04-11

服务器档案

文档知识
字段来自项目仓库与公开索引,中文说明由本站标注
提供方sonesuke上游标注的维护方 sonesuke
分类文档知识本站按上游分类展示
连接方式未标注上游未给出传输方式
鉴权方式会话登录远程服务建议用最小权限凭据
维护方社区 / 第三方核验等级 仅资料核对
包名 / 运行时未标注 · python本地安装时用到的包标识与运行时
部署类型未标注上游给出的部署形态
最近核验未标注上游最后核验时间

详细介绍

13.9 KB
来自项目仓库与上游目录的原文,未做删改

Oboyu (覚ゆ)

> ドキュメントを知識に、知識を価値に変える日本語特化型インテリジェンス・プラットフォーム
> Transform documents into knowledge, knowledge into value - Japanese-optimized Intelligence Platform

What is Oboyu?

Oboyu (覚ゆ - "to remember" in ancient Japanese) is a comprehensive Knowledge Intelligence Platform that transforms your documents into actionable insights. Going beyond traditional RAG (Retrieval-Augmented Generation), Oboyu combines advanced semantic search, knowledge graph generation, and AI-powered data enrichment to unlock the full potential of your information assets.

Beyond Traditional RAG

While most solutions stop at document retrieval, Oboyu creates a living knowledge ecosystem:
- Knowledge Graph Generation: Automatically extracts entities, relationships, and concepts from your documents
- GraphRAG Search: Leverages knowledge graphs for deeper, more contextual search results
- Data Enrichment: Enhances CSV files and structured data with insights from your knowledge base
- Multi-dimensional Intelligence: Combines vector search, graph traversal, and semantic analysis

Why Oboyu?

  • 🧠 Knowledge Intelligence: Automatically generates knowledge graphs and extracts insights from your documents
  • 📊 Data Enrichment: Enhances CSV files and structured data with AI-powered content from your knowledge base
  • 🚀 Lightning Fast: Indexes thousands of documents in seconds, searches in milliseconds with GraphRAG acceleration
  • 🎯 Beyond Accurate: Multi-layered search combining semantic understanding, knowledge graphs, and contextual reasoning
  • 🇯🇵 Japanese Excellence: Built specifically for Japanese business environments with automatic encoding detection
  • 🔒 Enterprise Private: Everything runs locally - your sensitive documents never leave your infrastructure
  • 🤖 AI-Native: Built-in MCP server for Claude, Cursor, and other AI assistants with GraphRAG capabilities

Quick Start

Prerequisites

  • Python 3.13 or higher (3.11+ supported)
  • pip (latest version recommended)
  • Operating System: Linux, macOS, or Windows with WSL

System Dependencies (for building from source)

Linux (Ubuntu/Debian):```bash
sudo apt-get install -y \
git \
curl \
build-essential \
cmake \
pkg-config \
libfreetype6-dev \
libfontconfig1-dev \
libjpeg-dev \
libpng-dev \
zlib1g-dev \
libssl-dev


**Linux (CentOS/RHEL):**```bash
sudo yum install -y \
    git \
    curl \
    gcc-c++ \
    cmake \
    pkg-config \
    freetype-devel \
    fontconfig-devel \
    libjpeg-devel \
    libpng-devel \
    zlib-devel \
    openssl-devel

macOS:```bash

Install Xcode Command Line Tools

xcode-select --install

Install additional dependencies via Homebrew

brew install cmake pkg-config


### Installation

Get up and running in under 5 minutes:```bash
# Install Oboyu
pip install oboyu

# Index your documents
oboyu index ~/Documents

# Search your documents
oboyu search "your search term"

That's it! See our Documentation for complete guides and examples.

Key Features

🧠 Knowledge Intelligence

  • Automatic Knowledge Graph Generation: Extracts entities, relationships, and concepts from your documents
  • GraphRAG Search: Leverages knowledge graphs for deeper, contextual search results
  • Multi-dimensional Associations: Discovers hidden connections between documents and concepts
  • Semantic Entity Recognition: Identifies and links key entities across your knowledge base
  • Relationship Mapping: Automatically maps relationships between concepts, people, and ideas

📊 Data Enrichment & Enhancement

  • CSV Auto-Enhancement: Enriches CSV files with relevant information from your knowledge base
  • Schema-Driven Processing: Uses JSON schema to define enrichment rules and data transformation
  • Semantic Data Completion: Fills missing information using AI-powered content matching
  • Business Value Creation: Transforms raw data into actionable business insights
  • Batch Processing: Efficiently processes large datasets with configurable batch sizes

🔍 Advanced Search Capabilities

  • Hybrid Search: Combines semantic understanding with keyword matching and graph traversal
  • Multiple Search Modes: Vector search, keyword search, GraphRAG, and hybrid modes
  • AI-Powered Reranking: Built-in reranker improves result accuracy and relevance
  • Contextual Understanding: Uses knowledge graphs to provide more relevant results
  • Flexible Output: Command-line search with JSON, plain text, and structured formats

📚 Comprehensive Document Support

  • Rich Format Support: PDF, plain text (.txt), Markdown (.md), HTML (.html), and source code files
  • PDF Intelligence: Advanced text extraction with metadata preservation and structure understanding
  • Incremental Indexing: Only processes new or changed files for lightning-fast updates
  • Smart Chunking: Intelligent document splitting optimized for knowledge extraction
  • Automatic Encoding: Seamlessly handles UTF-8, Shift-JIS, EUC-JP, and other encodings

🇯🇵 Japanese Business Excellence

  • Native Japanese Support: Purpose-built for Japanese business environments and content
  • Automatic Encoding Detection: Handles legacy Japanese encodings (Shift-JIS, EUC-JP) automatically
  • Specialized Language Models: Optimized embedding and processing models for Japanese text
  • Mixed Language Intelligence: Seamlessly processes Japanese-English bilingual documents
  • Business Context Understanding: Trained on Japanese business terminology and concepts

🚀 Enterprise Performance & Integration

  • ONNX Acceleration: 2-4x faster processing with automatic model optimization
  • MCP Server Integration: Native support for Claude Desktop and AI coding assistants
  • GraphRAG API: RESTful API for knowledge graph queries and data enrichment
  • Rich CLI Interface: Beautiful terminal interface with real-time progress tracking
  • Resource Efficient: Low memory footprint suitable for edge computing and local deployment

Installation

Using UV (Recommended)```bash

uv tool install oboyu


### Using pip```bash
pip install oboyu

From Source```bash

git clone https://github.com/sonesuke/oboyu.git
cd oboyu
pip install -e .


### System Requirements

- **Python**: 3.13 or higher (3.11+ supported)
- **OS**: macOS, Linux (Windows via WSL)
- **Memory**: 2GB RAM minimum (4GB recommended)
- **Storage**: 1GB for models and index
- **Build Tools**: See system dependencies above if building from source

> **Note**: Models are automatically downloaded on first use (~90MB).
> For installation from PyPI, most system dependencies are not required as we provide pre-built wheels.

## Usage Examples

### Basic Usage```bash
# Index a directory
oboyu index ~/Documents/notes

# Search your documents
oboyu search "machine learning optimization techniques"

# Get results in JSON format for processing
oboyu search "machine learning" --format json

Knowledge Intelligence & GraphRAG```bash

Build knowledge graph from your documents

oboyu build-kg

Search using GraphRAG for deeper insights

oboyu search "project management methodologies" --mode graphrag

Find related concepts and entities

oboyu search "agile development" --rerank --max-results 10


### Data Enrichment Workflows

**Schema Configuration (`enrichment_schema.json`):**```json
{
  "input_schema": {
    "columns": {
      "company_name": {"type": "string", "description": "Company name"}
    }
  },
  "enrichment_schema": {
    "columns": {
      "description": {
        "type": "string",
        "source_strategy": "search_content",
        "query_template": "{company_name} company overview business model"
      },
      "industry": {
        "type": "string",
        "source_strategy": "search_content",
        "query_template": "{company_name} industry sector business domain"
      }
    }
  }
}

Enrichment Commands:```bash

Enrich CSV with knowledge from your documents

oboyu enrich companies.csv enrichment_schema.json

Custom output location and batch processing

oboyu enrich data.csv schema.json -o enriched_data.csv --batch-size 5

Disable GraphRAG for faster processing

oboyu enrich simple_data.csv schema.json --no-graph


### Advanced Search Examples```bash
# Index only specific file types
oboyu index ~/projects --include-patterns "*.md,*.txt,*.pdf"

# GraphRAG search with relationship traversal
oboyu search "API design patterns" --mode graphrag --confidence 0.7

# Hybrid search combining multiple approaches
oboyu search "microservices architecture" --mode hybrid --rerank

# Search with custom result limits and confidence
oboyu search "database optimization" --max-results 15 --confidence 0.6

MCP Server for AI Assistants```bash

Start MCP server with GraphRAG capabilities

oboyu mcp

Or configure in Claude Desktop's settings


See our MCP Integration Guide for detailed setup instructions.

## Documentation

### 🚀 Getting Started
- **Installation** - Install and verify setup
- **Your First Index** - Create your first searchable index
- **Your First Search** - Learn to search effectively

### 💼 Real-world Usage
- **Daily Workflows** - Essential daily patterns
- **Technical Documentation** - Code and API docs
- **Meeting Notes** - Track decisions and actions
- **Research Papers** - Academic content search

### ⚙️ Configuration & Optimization
- **Configuration Guide** - Customize for your needs
- **Performance Tuning** - Optimize speed and quality
- **Japanese Support** - Japanese language features

### 🔗 Integration & Reference
- **Claude MCP Integration** - AI-powered search
- **CLI Reference** - All commands and options
- **Troubleshooting** - Solutions to common issues

**📖 View Full Documentation →**

## 🛠️ Technology Stack

Learn about the cutting-edge technologies that power Oboyu's intelligence:

- **📚 Technology Stack Overview** - Complete stack architecture and philosophy
- **🗄️ DuckDB: The Analytics Engine** - Why DuckDB powers our knowledge intelligence
- **🤖 HuggingFace: Japanese AI Excellence** - Specialized Japanese language models and embeddings
- **🔗 GraphRAG: Beyond Simple RAG** - Graph-enhanced retrieval and knowledge understanding
- **⚡ ONNX: Optimization Without Compromise** - 3x faster inference with maintained quality
- **⚖️ Our Decision Framework** - How we evaluate and choose technologies

We believe in transparency and sharing our technical journey. These deep-dives include performance benchmarks, implementation insights, and honest assessments of alternatives.

## Common Use Cases

### 🏢 Enterprise Knowledge Management
Transform organizational documents into a searchable knowledge graph:```bash
# Index company documents and build knowledge graph
oboyu index ~/company_docs --include "*.pdf,*.md,*.docx"
oboyu build-kg

# Search for strategic insights
oboyu search "competitive analysis market positioning" --mode graphrag

📊 Business Data Enhancement

Enrich customer or product data with insights from your knowledge base:```bash

Enhance customer list with company information

oboyu enrich customers.csv customer_enrichment_schema.json

Add product descriptions from documentation

oboyu enrich products.csv product_schema.json --batch-size 10


### 📚 Research & Academic Intelligence
Create a comprehensive research knowledge base:```bash
# Index research papers and notes
oboyu index ~/research --include "*.pdf,*.md,*.txt"
oboyu build-kg

# Find related concepts and methodologies
oboyu search "neural network optimization techniques" --mode graphrag

💻 Technical Documentation Intelligence

Make your codebase and documentation more discoverable:```bash

Index code and documentation

oboyu index ~/projects/myapp --include ".md,.py,.js,.java"

Find implementation patterns and examples

oboyu search "authentication middleware patterns" --rerank


### 📋 Meeting & Decision Intelligence
Transform meeting notes into actionable insights:```bash
# Index meeting notes and decisions
oboyu index ~/meetings --include "*.md,*.txt"

# Search for decisions and action items
oboyu search "budget approval Q4 initiatives" --mode hybrid

🌏 Multilingual Business Operations

Perfect for Japanese-English business environments:```bash

Index multilingual business documents

oboyu index ~/business_docs --include ".pdf,.md"

Search across languages seamlessly

oboyu search "プロジェクト管理 project management methodology" --mode graphrag


## Testing

### Unit and Integration Tests```bash
# Run fast tests (recommended for development)
uv run pytest -m "not slow"

# Run all tests with coverage
uv run pytest --cov=src

E2E Display Testing

Oboyu includes comprehensive E2E display testing using Claude Code SDK:```bash

Run all E2E display tests

python e2e/run_tests.py

Run specific test category

python e2e/run_tests.py --test search


See our Full Documentation for more details.

## Contributing

We welcome contributions! See our Contributing Guidelines for details.```bash
# Quick start for contributors
git clone https://github.com/YOUR_USERNAME/oboyu.git
cd oboyu
uv sync
uv run pytest -m "not slow"

Support

  • 📋 GitHub Issues - Report bugs or request features
  • 📖 Documentation - Comprehensive guides and references
  • 💬 Discussions - Ask questions and share ideas

License

This project is licensed under the MIT License - see the LICENSE.md file for details.

Acknowledgments

  • The name "Oboyu" (覚ゆ) comes from ancient Japanese, meaning "to remember"
  • Built with ❤️ for the Japanese business and NLP community
  • Inspired by the goal of making knowledge accessible and actionable across languages
  • Special thanks to the TinySwallow model for Japanese language understanding and knowledge extraction
  • GraphRAG implementation inspired by Microsoft's GraphRAG research and methodology

Made with 🇯🇵 by sonesuke

常见问题

7 问
答案由本站依据条目字段整理,不引用上游原文
Oboyu (覚ゆ) 怎么安装?

在客户端按下面的安装命令接入:pip install oboyu。装完在客户端配置里引用同一个包或端点。

Oboyu (覚ゆ) 怎么在 Claude Desktop、Claude、Cursor 里配置?

它的传输方式未标注,在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,配置才会生效。

Oboyu (覚ゆ) 收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于服务本身和背后的接口额度,正式使用前按官方说明确认。

Oboyu (覚ゆ) 需要 API Key 吗?

资料标注需要会话登录认证,使用前准备好对应凭据,并按最小权限配置。

Oboyu (覚ゆ) 支持哪些客户端?

资料标注支持 Claude Desktop、Claude、Cursor、Codex、Zed;其他客户端只要能配置 MCP Server,也能用同样的命令或地址接入。

Oboyu (覚ゆ) 能提供哪些能力?

资料显示能力标签包括 python、search engine、information retrieval、japanese、embeddings、semantic search。具体工具清单以官方文档为准。

Oboyu (覚ゆ) 需要什么运行环境?

运行时为 python。

热门专题

按标签挑服务器
全部专题 →

服务器信息以作者仓库与官方文档为准;分类、传输方式与鉴权方式的中文说明为中文整理,最近核验 未标注。