Semantic Kernel Plugins

A collection of ready-to-use plugins for Microsoft's Semantic Kernel framework that enhances AI applications with database connectivity, shell operations, web search capabilities, Python code execution, and detailed logging.

🚀 Overview

Semantic Kernel Plugins provides a set of powerful, production-ready plugins for the Semantic Kernel Framework, eliminating the need to write your own plugins from scratch:

• PostgreSQL Plugin: Connect and interact with PostgreSQL databases with ease
• MongoDB Plugin: Use MongoDB databases directly in your AI workflows
• Shell Plugin: Execute system commands safely across different operating systems
• Web Search Plugins: Choose from multiple search providers (Tavily, Google, SerpAPI)
• Python Code Generator: Generate and execute Python code safely

✨ Features

• Multiple Database Integrations: Connect to PostgreSQL and MongoDB databases
• Cross-Platform Shell Operations: Run system commands with proper handling for Windows, Linux, and macOS
• Multi-Provider Web Search: Choose from Tavily, Google Search, or SerpAPI for web searches
• Python Code Generation and Execution: Generate and execute Python code in a controlled environment
• Enhanced Logging: Track operations with detailed, colorful logs
• Cross-Platform Compatibility: Works seamlessly across Windows, Linux, and macOS

📋 Requirements

• Python 3.10+
• Semantic Kernel 1.0.0+

📥 Installation

pip install semantic-kernel-plugins

🚀 Quick Start

Database Plugins

PostgreSQL Plugin

import os
from dotenv import load_dotenv
from semantic_kernel import Kernel
from psycopg2 import connect
from semantic_kernel_plugins.plugins.postgre import PostgrePlugin

# Initialize connection
db_connection = connect(
    host=os.getenv("DB_HOST"),
    database=os.getenv("DB_NAME"),
    user=os.getenv("DB_USER"),
    password=os.getenv("DB_PASSWORD")
)

# Create kernel and add plugin
kernel = Kernel()
kernel.add_plugin(
    PostgrePlugin(db_connection),
    plugin_name="PostgreSQL"
)

# Example usage
# Execute SQL queries directly from your AI workflows
result = kernel.plugins["PostgreSQL"].execute_query("SELECT * FROM users")

# Get table schemas
tables = kernel.plugins["PostgreSQL"].fetch_table_names()
schema = kernel.plugins["PostgreSQL"].fetch_table_schema("users")

MongoDB Plugin

from pymongo import MongoClient
from semantic_kernel import Kernel
from semantic_kernel_plugins.plugins.mongodb import MongoDBPlugin

# Initialize MongoDB client
client = MongoClient("mongodb://localhost:27017/")

# Create kernel and add plugin
kernel = Kernel()
kernel.add_plugin(
    MongoDBPlugin(client),
    plugin_name="MongoDB"
)

# Example usage
# Check if database exists
exists = kernel.plugins["MongoDB"].database_exists("mydatabase")

# List all collections in a database
collections = kernel.plugins["MongoDB"].list_collections("mydatabase")

# Get database statistics
stats = kernel.plugins["MongoDB"].get_database_stats("mydatabase")

Shell Plugin

from semantic_kernel import Kernel
from semantic_kernel_plugins.plugins.shell import ShellPlugin

# Create kernel and add shell plugin
kernel = Kernel()
kernel.add_plugin(
    ShellPlugin(),
    plugin_name="Shell"
)

# Execute shell commands safely across platforms
result = kernel.plugins["Shell"].execute_shell_command("ls -la")

# You can also pass a list of arguments
result = kernel.plugins["Shell"].execute_shell_command(["python", "-c", "print('Hello from Python!')"])

Web Search Plugins

Tavily Search Plugin

import os
from semantic_kernel import Kernel
from semantic_kernel_plugins.plugins.web.tavily_web_search import TavilySearchPlugin
from semantic_kernel_plugins.logger.sk_logger import SKLogger

# Setup logger for detailed search result tracking
logger = SKLogger(name="SearchApp")

# Create kernel and add web search plugin
kernel = Kernel()
kernel.add_plugin(
    TavilySearchPlugin(
        api_key=os.getenv("TAVILY_API_KEY"),
        search_depth="advanced",
        include_answer=True,
        logger=logger
    ),
    plugin_name="TavilySearch"
)

# Search the web with detailed results
results = kernel.plugins["TavilySearch"].search("latest advancements in quantum computing")

SerpAPI Search Plugin

import os
from semantic_kernel import Kernel
from semantic_kernel_plugins.plugins.web import SerpApiWebSearchPlugin

# Create kernel and add SerpAPI search plugin
kernel = Kernel()
kernel.add_plugin(
    SerpApiWebSearchPlugin(
        api_key=os.getenv("SERPAPI_API_KEY"),
        engine="google",
        include_news=True  # Get additional news results
    ),
    plugin_name="SerpApiSearch"
)

# Search the web with Google via SerpAPI
results = kernel.plugins["SerpApiSearch"].search("current global economic trends")

Google Search Plugin

from semantic_kernel import Kernel
from semantic_kernel_plugins.plugins.web import GoogleSearchPlugin

# Create kernel and add Google search plugin
kernel = Kernel()
kernel.add_plugin(
    GoogleSearchPlugin(
        max_results=10,
        advanced=True
    ),
    plugin_name="GoogleSearch"
)

# Search the web directly with Google
results = kernel.plugins["GoogleSearch"].google_search("best practices for cloud security")

📦 Available Plugins

Plugin	Description
Database Plugins
PostgrePlugin	Execute queries and manage PostgreSQL databases
MongoDBPlugin	Interact with MongoDB databases and collections
System Plugins
ShellPlugin	Execute shell commands across different platforms safely
Web Search Plugins
TavilySearchPlugin	AI-powered search with summarization via Tavily
SerpApiWebSearchPlugin	Comprehensive search results via SerpAPI (Google)
GoogleSearchPlugin	Direct Google search integration
Development Plugins
PythonCodeGeneratorPlugin	Generate and execute Python code safely
CalculatorPlugin	Perform mathematical calculations

🔍 Detailed Plugin Features

PostgreSQL Plugin

• Execute arbitrary SQL queries
• Fetch table names and schemas
• Insert, update, and delete data
• Create and drop tables

MongoDB Plugin

• List databases and collections
• Check if databases or collections exist
• Get database and collection statistics

Shell Plugin

• Cross-platform command execution
• Safe handling of command arguments
• Proper error reporting and logging

Web Search Plugins

• Multiple search providers for different needs
• Configurable result formatting (markdown or JSON)
• Rich search results with titles, snippets, and URLs
• Advanced options for specialized searches (news, images, etc.)

🔜 Coming Soon

• .NET Plugins
• SQLite Plugin
• File System Plugin
• More Cloud Service Integrations

📄 License

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

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

---

Built with ❤️ for AI developers working with Microsoft's Semantic Kernel.