Agentium - AI Agent Development Toolkit

A comprehensive Python library designed for AI agent development and workflow orchestration. Agentium provides a rich set of tools and utilities that seamlessly integrate with popular AI frameworks like LangChain, LangGraph, and now features built-in Google Gemini API integration.

New Features

• Google Gemini Integration - Built-in support for Gemini Pro, Gemini 1.5 Pro, and Gemini 1.5 Flash
• Model Selection - Easy switching between different AI models
• Enhanced AI Processing - All core features now support AI-enhanced processing
• Advanced Configuration - Comprehensive model and processing configuration

Features

• Condense: Intelligent content condensation and compression with AI enhancement
• Optimizer: Refine text, code, and workflows for better performance
• Rearranger: Organize and restructure content logically
• Extractor: Extract structured information from various data sources
• Communicator: Send messages and notifications across platforms
• Translator: Multi-language translation with AI-powered tone adaptation
• Insight Generator: Generate actionable insights from data using AI
• Workflow Helper: Orchestrate complex tasks and triggers
• Template Manager: Standardize outputs with customizable templates
• Memory Helper: Context storage and retrieval system
• Custom Summarizer: Create AI-enhanced summaries tailored to specific needs
• Logger Utils: Track and debug operations with detailed logging
• Gemini Integration: Native Google Gemini API support with model selection

Installation

pip install agentium

With AI Enhancement (Recommended):

# For full AI capabilities including Gemini
pip install agentium google-generativeai

# For LangChain integration
pip install agentium[langchain]

# For LangGraph integration
pip install agentium[langgraph]

# For development
pip install agentium[dev]

API Key Setup

Getting Your Google Gemini API Key

To use the AI-enhanced features, you need a Google Gemini API key:

1. Visit Google AI Studio

   • Go to https://makersuite.google.com/app/apikey
   • Sign in with your Google account

2. Create API Key

   • Click "Create API Key"
   • Choose "Create API key in new project" or select an existing project
   • Copy the generated API key and save it securely

3. Set Up Your API Key (Choose one method):

   Method 1: Environment Variable (Recommended)

   # Windows (Command Prompt)
   set GEMINI_API_KEY=your_api_key_here
   
   # Windows (PowerShell)
   $env:GEMINI_API_KEY="your_api_key_here"
   
   # Linux/Mac
   export GEMINI_API_KEY=your_api_key_here
   

   Method 2: In Your Python Code

   from agentium.integrations.gemini import GeminiIntegration
   
   # Initialize with API key directly
   gemini = GeminiIntegration(api_key="your_api_key_here")
   

   Method 3: Using .env File

   # Create a .env file in your project root
   echo "GEMINI_API_KEY=your_api_key_here" > .env
   

   import os
   from dotenv import load_dotenv
   
   load_dotenv()
   api_key = os.getenv('GEMINI_API_KEY')
   

Important Security Notes

• Never commit API keys to version control
• Use environment variables in production
• Keep your API keys secure and private
• Monitor your API usage in Google Cloud Console

Quick Start

from agentium import Agentium
from agentium.integrations.gemini import GeminiIntegration

# Initialize with Gemini AI enhancement
agentium = Agentium()
gemini = GeminiIntegration(api_key="your-gemini-api-key")

# Basic text processing
condensed_text = agentium.condenser.condense("Your long text here...")
optimized_text = agentium.optimizer.optimize(condensed_text)

# AI-enhanced processing
enhanced_summary = gemini.enhance_condenser(
    text="Complex document content",
    style="executive-summary"
)

# Workflow orchestration
workflow_result = agentium.workflow_helper.process_workflow({
    "input": "Data to process",
    "steps": ["condense", "optimize", "extract"],
    "ai_enhanced": True
})

print(f"Processed result: {workflow_result}")

Configuration with Gemini

from agentium.integrations.gemini import GeminiIntegration, GeminiConfig

# Configure Gemini integration
config = GeminiConfig(
    model_name="gemini-1.5-pro",
    temperature=0.7,
    max_tokens=1000
)

gemini = GeminiIntegration(
    api_key="your-api-key",
    config=config
)

# Use with any Agentium feature
enhanced_insights = gemini.enhance_insights(
    data="Your data here",
    context="Financial analysis",
    insight_type="trend_analysis"
)

Sample Projects

The sample_projects/ directory contains 5 comprehensive examples demonstrating all Agentium features:

1. Content Processing Pipeline (content_pipeline.py)

Advanced document processing system with AI-enhanced content analysis, multi-format support, and intelligent workflow orchestration.

Features Demonstrated:

• Content condensation and optimization
• Multi-language translation with tone adaptation
• Template-based report generation
• Memory management and context storage
• Gemini AI enhancement for superior content quality

# Example usage
pipeline = ContentProcessingPipeline()
result = pipeline.process_document("document.pdf", output_format="executive-summary")

2. Multilingual News Analyzer (news_analyzer.py)

Real-time news analysis system with sentiment analysis, trend detection, and multi-language support.

Features Demonstrated:

• Real-time content extraction and processing
• Advanced insight generation with trend analysis
• Multi-language translation and sentiment analysis
• Workflow orchestration for news processing
• AI-enhanced content understanding

# Example usage
analyzer = NewsAnalyzer()
analysis = analyzer.analyze_news_feed("https://news-feed-url", languages=["en", "es", "fr"])

3. Data Intelligence Dashboard (data_dashboard.py)

Comprehensive data analysis and visualization system with AI-powered insights.

Features Demonstrated:

• Advanced data extraction and processing
• AI-enhanced insight generation
• Real-time data visualization
• Memory-based context management
• Intelligent report generation

# Example usage
dashboard = DataIntelligenceDashboard()
insights = dashboard.generate_intelligence_report("sales_data.csv")

4. Automated Report Generator (report_generator.py)

Professional report generation system with AI-enhanced content creation and multi-format output.

Features Demonstrated:

• Template-based report generation
• AI-enhanced content optimization
• Multi-format output (PDF, HTML, Markdown)
• Workflow automation for report creation
• Memory management for report templates

# Example usage
generator = AutomatedReportGenerator()
report = generator.generate_comprehensive_report("project_data", "quarterly-report")

5. Smart Communication Hub (communication_hub.py)

Intelligent communication orchestration system with multi-channel messaging and workflow automation.

Features Demonstrated:

• Multi-channel communication management
• Workflow-based message orchestration
• AI-enhanced content optimization for different channels
• Memory management for communication history
• Intelligent message routing and optimization

# Example usage
hub = SmartCommunicationHub()
result = hub.process_and_distribute_message(
    "Important announcement",
    channels=["email", "slack", "teams"],
    ai_enhanced=True
)

Streamlit Demo

Try the interactive demo to explore all Agentium features:

# Install Streamlit
pip install streamlit plotly pandas

# Run the demo
streamlit run sample_projects/agentium_streamlit_demo.py

The demo includes:

• Interactive feature testing
• Real-time processing examples
• AI model selection interface
• Processing history and export capabilities
• Comprehensive feature demonstrations

Framework Integration

LangChain Integration

from agentium.integrations.langchain import AgentiumTool
from langchain.agents import initialize_agent

# Create Agentium tools for LangChain
tools = AgentiumTool.create_all_tools()
agent = initialize_agent(tools, llm, agent_type="zero-shot-react-description")

LangGraph Integration

from agentium.integrations.langgraph import AgentiumNode
from langgraph import Graph

# Add Agentium nodes to LangGraph
graph = Graph()
graph.add_node("condenser", AgentiumNode.condenser_node)
graph.add_node("optimizer", AgentiumNode.optimizer_node)

Documentation

For detailed documentation and examples, visit our documentation site.

Contributing

We welcome contributions! Please see our Contributing Guide for details.

License

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

Support

If you encounter any issues or have questions, please open an issue on GitHub.