Event Management AI Agent

A powerful AI-powered event scheduling and calendar management system that processes natural language requests to create, store, and manage events. Built with LangChain, OpenAI GPT, and Python.

Features

• Natural Language Processing: Schedule events using plain English
• Dual Storage: Events saved in both ICS calendar format and CSV backup
• Smart Extraction: AI-powered extraction of event details from text
• Flexible Queries: List upcoming events with various time filters
• Interactive CLI: Easy-to-use command-line interface
• Robust Fallback: Fallback parsing when AI extraction fails

Installation

1. Clone the repository:

git clone https://github.com/avinash00134/event-management-agent.git
cd event-management-agent

2. Install required dependencies:

pip install langchain langchain-openai pydantic ics pathlib

3. Set up your OpenAI API key:

export OPENAI_API_KEY='your-api-key-here'

Usage

Command Line Interface

Run the interactive agent:

python event_agent.py

Example Commands

The agent understands natural language requests like:

• "Schedule a meeting with John tomorrow at 2pm"
• "Add dentist appointment on July 25 at 9:30 AM at Downtown Clinic"
• "Team standup every Monday at 10am in Conference Room A"
• "Show my upcoming events"
• "List events for next week"
• "What do I have today?"

Programmatic Usage

from event_agent import EventManagementAgent

# Initialize the agent
agent = EventManagementAgent(openai_api_key="your-api-key")

# Process event requests
result = agent.process_event_request("Schedule lunch with Sarah tomorrow at 1pm")
print(result)

# List upcoming events
events = agent.process_event_request("Show my events for next week")
print(events)

Architecture

Core Components

1. EventDetails: Pydantic model for structured event data
2. EventStorage: Handles dual storage in ICS and CSV formats
3. EventManagementAgent: Main agent orchestrating the workflow
4. Extraction Tools: AI-powered natural language processing

Storage Files

• events.ics: Standard ICS calendar file for calendar applications
• events_backup.csv: CSV backup with full event details
• event_log.txt: Activity log for debugging and tracking

Available Tools

• extract_and_add_event: Extract and immediately add events
• extract_event_details_only: Preview extraction without adding
• list_upcoming_events: Query events within specified timeframes

Event Data Structure

Events are structured with the following fields:

{
    "title": str,           # Event name/title
    "date": str,            # YYYY-MM-DD format
    "time": str,            # HH:MM format (24-hour)
    "location": str,        # Optional venue/location
    "attendees": List[str], # Optional list of attendee names
    "description": str,     # Optional additional details
    "created_at": str       # Timestamp of creation
}

Natural Language Processing

The agent can interpret various time and date formats:

• Relative dates: "tomorrow", "next Monday", "in 3 days"
• Time formats: "2pm", "14:30", "9:00 AM"
• Location extraction: "at Downtown Office", "in Room 101"
• Attendee parsing: "with John and Sarah", "meeting with team"

Configuration

Key configuration constants in the code:

CALENDAR_FILE = "events.ics"      # ICS calendar output
CSV_FILE = "events_backup.csv"    # CSV backup file
EVENT_LOG = "event_log.txt"       # Activity log

Error Handling

The system includes robust error handling:

• Fallback parsing when AI extraction fails
• File creation if storage files don't exist
• Graceful degradation for malformed inputs
• Detailed error messages for debugging

Requirements

• Python 3.7+
• OpenAI API key
• Required packages:
  • langchain
  • langchain-openai
  • pydantic
  • ics
  • pathlib

File Structure

event-management-agent/
├── event_agent.py          # Main application file
├── events.ics             # Generated ICS calendar
├── events_backup.csv      # CSV backup storage
├── event_log.txt          # Activity log
└── README.md              # This file

Examples

Adding Events

# Simple event
"Lunch tomorrow at noon"

# Detailed event
"Schedule project review on July 30th at 3pm in Conference Room B with Alice, Bob, and Charlie"

# Recurring context
"Weekly team meeting every Thursday at 10am"

Querying Events

# Time-based queries
"Show today's events"
"What's coming up this week?"
"List events for the next month"

# General queries
"Show my calendar"
"Upcoming events"
"What do I have scheduled?"

Test Suite

The test suite (test_event_agent.py) verifies the core functionality of the Event Management AI Agent with the following test cases:

Event Extraction Tests

1. Basic Extraction

   • Verifies extraction of complete event details (title, date, time, location, attendees)
   • Example: "Schedule a meeting with Alice tomorrow at 3 PM in Conference Room A"

2. Missing Time Handling

   • Tests default time assignment when not specified
   • Example: "Book a yoga class next Monday" → Defaults to 12:00

3. Today/Tomorrow Handling

   • Validates relative date calculations
   • Example: "Set up a quick sync today" → Uses current date

4. Event Validation

   • Ensures extracted data passes Pydantic model validation
   • Example: "Team meeting Friday at 2pm"

Tool Functionality Tests

5. Extract and Add Tool

   • Tests the end-to-end tool that extracts and stores events
   • Example: "Doctor appointment tomorrow at 3pm"

6. Extract Only Tool

   • Verifies the preview functionality without storage
   • Example: "Team sync next Monday at 9 AM"

Agent Integration Tests

7. Event Creation Flow

   • Tests the full agent workflow for creating events
   • Mocks both extraction and storage components

8. Event Listing Flow

   • Verifies the agent's ability to list upcoming events
   • Example: "List events for next week"

Running Tests

python -m pytest test_event_agent.py -v