Message Steering Protocol • AI Agent SDK

MessageSteeringProtocol: Protocol for message steering during agent execution.

MessageSteeringProtocol

Defined in the protocols module.

AI Agent

Protocol for message steering during agent execution.

This protocol enables real-time communication with agents during long-running operations, allowing users to provide mid-execution guidance, feedback, and course corrections.

Methods

queue_message()

Queue a steering message for the agent.

get_pending_messages()

Get all pending steering messages.

process_steering()

Process any pending steering messages and update execution context.

clear_messages()

Clear all pending messages.

has_pending_messages()

Check if there are pending steering messages.

enabled()

Whether message steering is enabled.

Usage

python

class BasicMessageSteering:
        def queue_message(self, message: str, priority: int = 5) -> str:
            # Add to internal queue
            return "msg_001"

        def get_pending_messages(self) -> List[SteeringMessage]:
            # Return queued messages
            return []

        def process_steering(self, context: dict) -> bool:
            # Check for messages and update context
            return False

    agent = Agent(name="assistant", message_steering=BasicMessageSteering())

Source

View on GitHub

praisonaiagents/agent/protocols.py at line 454