aliyun/rdsai-cli

RDSAI CLI is a next-generation, AI-powered RDS CLI that transforms how you interact with the databasea

62

stars

147

commits

Python

primary language

May 2, 2026

updated

agent
agentic
ai
cli
python

README

RDSAI CLI

AI-Powered Database Command-Line Interface

Python 3.13+ License GitHub issues

English | 中文

RDSAI CLI Screenshot

RDSAI CLI is an AI-powered command-line interface for database management.
It supports natural language queries and SQL execution, with built-in diagnostic tools,
execution plan analysis, and multi-source data connections.


Features

  • Multi-Source Connection - Connect to MySQL databases or files (CSV, Excel) locally or remotely via HTTP/HTTPS
  • AI Assistant - Natural language queries (English/Chinese), SQL optimization, diagnostics, and explanations
  • Smart SQL - Auto-detects SQL vs natural language, query history, Ctrl+E for instant result explanations
  • Multi-Model LLM - Support for Qwen, OpenAI, DeepSeek, Anthropic, Gemini, and OpenAI-compatible APIs
  • Schema Analysis - AI-powered database analysis with compliance checking and optimization suggestions
  • Performance Benchmarking - Automated sysbench testing with comprehensive analysis reports
  • MCP Integration - Extend capabilities via Model Context Protocol servers
  • Safety First - Read-only by default, DDL/DML requires confirmation (YOLO mode available)

Requirements

  • Python 3.13+
  • Network access to MySQL
  • API access to LLM provider

Installation

# One-click installation (recommended)
curl -LsSf https://raw.githubusercontent.com/aliyun/rdsai-cli/main/install.sh | sh

# Or using uv
uv tool install --python 3.13 rdsai-cli

# Or using pip (recommend virtual environment)
pip install rdsai-cli

Quick Start

Launch and Connect

# Start without connection (interactive mode)
rdsai

# Connect via command line
rdsai --host localhost -u root -p secret -D mydb

# Connect to files
rdsai
> /connect flights.csv
> /connect https://example.com/data.csv

Configure LLM

mysql> /setup

Interactive wizard will guide you through LLM provider setup. Configuration is saved to ~/.rdsai-cli/config.json.

Basic Usage

SQL Execution:

mysql> SELECT COUNT(*) FROM users;
mysql> EXPLAIN SELECT * FROM users WHERE email = 'test@example.com';
mysql> SELECT * FROM users LIMIT 10\G   -- Press Ctrl+E for AI explanation

Natural Language:

mysql> analyze index usage on users table
mysql> show me slow queries from the last hour
mysql> design an orders table for e-commerce
mysql> why this query is slow: SELECT * FROM users WHERE name LIKE '%john%'

Meta Commands

CommandDescription
/connect, /disconnectConnect/disconnect from database or files
/setupConfigure LLM provider
/helpShow help and status
/explainAnalyze SQL execution plan
/researchGenerate database schema analysis report
/benchmarkRun performance benchmark test
/yoloToggle auto-approve mode (use with caution)
/historyShow query history
/modelManage LLM models

Documentation

  • Complete Tutorial - Comprehensive guide from beginner to advanced
  • Core Features - Execution plan analysis, schema analysis, benchmarking, MCP integration
  • Usage Scenarios - SQL explanation, file analysis, slow query optimization, and more

Security

  • Read-Only by Default - DDL/DML require explicit confirmation unless YOLO mode is enabled
  • Confirmation Required - Every write operation shows exact SQL for review before execution
  • Credential Storage - API keys stored in ~/.rdsai-cli/config.json (protect with proper permissions)

Contributing

We welcome contributions. See CONTRIBUTING.md for details.

License

MIT License - see LICENSE for details.

Contributors

wangype

143 commits

fdwills

4 commits

aliyun/rdsai-cli

RDSAI CLI is a next-generation, AI-powered RDS CLI that transforms how you interact with the databasea

62

stars

147

commits

Python

primary language

May 2, 2026

updated

agent
agentic
ai
cli
python

README

RDSAI CLI

AI-Powered Database Command-Line Interface

Python 3.13+ License GitHub issues

English | 中文

RDSAI CLI Screenshot

RDSAI CLI is an AI-powered command-line interface for database management.
It supports natural language queries and SQL execution, with built-in diagnostic tools,
execution plan analysis, and multi-source data connections.


Features

  • Multi-Source Connection - Connect to MySQL databases or files (CSV, Excel) locally or remotely via HTTP/HTTPS
  • AI Assistant - Natural language queries (English/Chinese), SQL optimization, diagnostics, and explanations
  • Smart SQL - Auto-detects SQL vs natural language, query history, Ctrl+E for instant result explanations
  • Multi-Model LLM - Support for Qwen, OpenAI, DeepSeek, Anthropic, Gemini, and OpenAI-compatible APIs
  • Schema Analysis - AI-powered database analysis with compliance checking and optimization suggestions
  • Performance Benchmarking - Automated sysbench testing with comprehensive analysis reports
  • MCP Integration - Extend capabilities via Model Context Protocol servers
  • Safety First - Read-only by default, DDL/DML requires confirmation (YOLO mode available)

Requirements

  • Python 3.13+
  • Network access to MySQL
  • API access to LLM provider

Installation

# One-click installation (recommended)
curl -LsSf https://raw.githubusercontent.com/aliyun/rdsai-cli/main/install.sh | sh

# Or using uv
uv tool install --python 3.13 rdsai-cli

# Or using pip (recommend virtual environment)
pip install rdsai-cli

Quick Start

Launch and Connect

# Start without connection (interactive mode)
rdsai

# Connect via command line
rdsai --host localhost -u root -p secret -D mydb

# Connect to files
rdsai
> /connect flights.csv
> /connect https://example.com/data.csv

Configure LLM

mysql> /setup

Interactive wizard will guide you through LLM provider setup. Configuration is saved to ~/.rdsai-cli/config.json.

Basic Usage

SQL Execution:

mysql> SELECT COUNT(*) FROM users;
mysql> EXPLAIN SELECT * FROM users WHERE email = 'test@example.com';
mysql> SELECT * FROM users LIMIT 10\G   -- Press Ctrl+E for AI explanation

Natural Language:

mysql> analyze index usage on users table
mysql> show me slow queries from the last hour
mysql> design an orders table for e-commerce
mysql> why this query is slow: SELECT * FROM users WHERE name LIKE '%john%'

Meta Commands

CommandDescription
/connect, /disconnectConnect/disconnect from database or files
/setupConfigure LLM provider
/helpShow help and status
/explainAnalyze SQL execution plan
/researchGenerate database schema analysis report
/benchmarkRun performance benchmark test
/yoloToggle auto-approve mode (use with caution)
/historyShow query history
/modelManage LLM models

Documentation

  • Complete Tutorial - Comprehensive guide from beginner to advanced
  • Core Features - Execution plan analysis, schema analysis, benchmarking, MCP integration
  • Usage Scenarios - SQL explanation, file analysis, slow query optimization, and more

Security

  • Read-Only by Default - DDL/DML require explicit confirmation unless YOLO mode is enabled
  • Confirmation Required - Every write operation shows exact SQL for review before execution
  • Credential Storage - API keys stored in ~/.rdsai-cli/config.json (protect with proper permissions)

Contributing

We welcome contributions. See CONTRIBUTING.md for details.

License

MIT License - see LICENSE for details.

Contributors

wangype

143 commits

fdwills

4 commits

Languages

Python

98.2%

Shell

1.8%