🚀 Ryoma AI Documentation¶

AI-Powered Data Analysis Platform Connect to databases, ask questions in natural language, and get intelligent insights

🎯 What is Ryoma?¶

Ryoma is a cutting-edge AI-powered data platform that revolutionizes how data users interact with their data. Built on state-of-the-art research, Ryoma enables:

🤖 Intelligent SQL Generation¶

Enhanced SQL Agent - Multi-step reasoning with safety validation
ReFoRCE Agent - Research-based self-refinement for maximum accuracy
Natural Language Queries - Ask questions in plain English, get SQL results

📊 Advanced Database Profiling¶

Comprehensive Metadata Extraction - Automatic schema understanding
Data Quality Assessment - Multi-dimensional quality scoring
Semantic Type Detection - Automatic identification of emails, phones, IDs
Column Similarity Analysis - LSH-based relationship discovery

🗄️ Universal Database Support¶

PostgreSQL, MySQL, Snowflake, BigQuery - Production-ready connectors
SQLite, DuckDB - Perfect for development and analytics
Ibis Integration - Native database optimizations for better performance

🛡️ Enterprise-Ready Security¶

Query Validation - Configurable safety policies
Access Control - Fine-grained permissions
Audit Logging - Complete query tracking

Ryoma Architecture

👥 Who is Ryoma for?¶

📈 Data Analysts¶

Transform natural language questions into complex SQL queries without deep SQL knowledge.

🔬 Data Scientists¶

Rapidly explore datasets and generate insights with AI-powered analysis.

💼 Business Users¶

Get answers from your data without waiting for technical teams.

🏢 Enterprise Teams¶

Deploy secure, scalable data analysis with comprehensive governance.

🚀 Quick Start¶

Get up and running in under 5 minutes:

from ryoma_ai.agent.sql import SqlAgent
from ryoma_ai.datasource.postgres import PostgresDataSource

# Connect to your database with profiling
datasource = PostgresDataSource(
    connection_string="postgresql://user:pass@localhost:5432/db",
    enable_profiling=True  # Automatic metadata extraction
)

# Create enhanced SQL agent
agent = SqlAgent(model="gpt-4", mode="enhanced")
agent.add_datasource(datasource)

# Ask questions in natural language
response = agent.stream("Show me the top 10 customers by revenue this quarter")
print(response)

🎯 Key Features¶

🚀 Feature	📝 Description	🔗 Learn More
Enhanced SQL Agent	Multi-step reasoning with safety validation	Agent Guide →
Database Profiling	Comprehensive metadata extraction	Profiling Guide →
Universal Connectors	Support for all major databases	Data Sources →
Safety Framework	Configurable validation and security	Advanced Setup →
Model Flexibility	OpenAI, Anthropic, local models	Models →