Communications Agent

This is a communications agent that can read and process messages from a shared thread with a scheduler agent in the same Azure AI Project.

• ✅ Read the last posted message from a shared thread
• ✅ Monitor a shared thread for new messages
• ✅ Process messages from the scheduler agent
• ✅ Send acknowledgments back to the thread
• ✅ Handle different message types and content

Setup

1. Make sure you have the required environment variables set:

   PROJECT_CONNECTION_STRING=your_azure_project_connection_string
   MODEL_DEPLOYMENT_NAME=your_model_deployment_name

2. Optional: Set the shared thread ID as an environment variable:

   SHARED_THREAD_ID=thread_abc123xyz

Usage

Quick Check - Read Last Message

Use the simple check script to quickly read the last message from a shared thread:

python check_messages.py

Make sure to set the THREAD_ID variable in the script or the SHARED_THREAD_ID environment variable.

Full Agent Functionality

Run the main agent with full functionality:

python main.py

Key Functions

read_last_message_from_thread(project_client, thread_id)

• Reads the most recent message from a thread
• Returns message details including content, role, and timestamp
• Handles different content types (text, files, etc.)

monitor_shared_thread(thread_id, check_interval)

• Continuously monitors a thread for new messages
• Processes messages from the scheduler agent
• Can send acknowledgments back to the thread
• Configurable check interval (default: 30 seconds)

process_scheduler_message(message_content, project_client, thread_id)

• Processes messages from the scheduler agent
• Detects scheduling-related keywords
• Can trigger notifications or other actions
• Sends acknowledgments back to the thread

Message Flow

1. Scheduler Agent posts a message to the shared thread
2. Communications Agent detects the new message
3. Communications Agent processes the message content
4. Communications Agent can:
   • Send notifications (email, SMS, Slack, etc.)
   • Update databases
   • Send acknowledgment back to the thread
   • Trigger other workflows

Example Usage

# Read the last message from a shared thread
last_message = await read_last_message_from_thread(project_client, "thread_abc123")

# Monitor a shared thread continuously
await monitor_shared_thread("thread_abc123", check_interval=10)

# Display recent messages from a thread
await utilities.display_thread_messages(project_client, "thread_abc123", limit=5)

Monitoring Setup

To continuously monitor a shared thread:

1. Set the existing_thread_id variable in main.py
2. Uncomment the monitor_shared_thread() call
3. Run the script - it will continuously check for new messages

Error Handling

The agent includes comprehensive error handling for:

• Connection issues
• Invalid thread IDs
• Message parsing errors
• Network timeouts

Customization

You can customize the agent by:

• Modifying the message processing logic in process_scheduler_message()
• Adding custom notification methods
• Changing the monitoring interval
• Adding filters for specific message types
• Implementing custom response templates

Integration Ideas

• Email Notifications: Send emails when schedule changes are detected
• SMS Alerts: Send text messages for urgent schedule updates
• Slack/Teams Integration: Post messages to communication channels
• Database Updates: Update scheduling databases with changes
• Calendar Integration: Sync with calendar applications
• Webhook Triggers: Trigger external systems via webhooks