Skip to main content
Inputs let you send additional data asynchronously to BASIC while an agent is running. They are useful when integrating external signals such as added sensors (for example directional microphone or air-quality sensor), network events (for example incoming emails or webhook alerts), and internal robot status events. The most important part is that an input can send live feedback to BASIC while the agent is executing.

​
Minimal input example (stripped down)

import threading
from brain_client.input_types import InputDevice

class AlertsInput(InputDevice):
    def __init__(self):
        super().__init__()
        self._stop = threading.Event()

    @property
    def name(self) -> str:
        return "alerts_input"

    def on_open(self):
        self._stop.clear()

        def loop():
            while not self._stop.is_set():
                events = self._poll_events()  # your sensor/API hook
                for event_text in events:
                    # Async feedback to BASIC during agent execution
                    self.send_data(event_text, data_type="chat_in")
                self._stop.wait(1.0)

        threading.Thread(target=loop, daemon=True).start()

    def on_close(self):
        self._stop.set()

    def _poll_events(self):
        return []

​
Attach it to an agent

def get_inputs(self) -> list[str]:
    return ["alerts_input"]
When this agent starts, on_open() is called and BASIC begins receiving these asynchronous updates. For more details: