Warning: This document is for the development version of Rasa. The latest version is 1.10.8.

Custom Connectors

You can also implement your own custom channel. You can use the rasa.core.channels.channel.RestInput class as a template. The methods you need to implement are blueprint and name. The method needs to create a sanic blueprint that can be attached to a sanic server.

This allows you to add REST endpoints to the server that the external messaging service can call to deliver messages.

Your blueprint should have at least the two routes: health on /, and receive on the HTTP route /webhook.

The name method defines the url prefix. E.g. if your component is named myio, the webhook you can use to attach the external service is: http://localhost:5005/webhooks/myio/webhook (replacing the hostname and port with your values).

To send a message, you would run a command like:

curl -XPOST http://localhost:5005/webhooks/myio/webhook \
  -d '{"sender": "user1", "message": "hello"}' \
  -H "Content-type: application/json"

where myio is the name of your component.

If you need to use extra information from your front end in your custom actions, you can add this information in the metadata dict of your user message. This information will accompany the user message through the rasa server into the action server when applicable, where you can find it stored in the tracker. Message metadata will not directly affect NLU classification or action prediction. If you want to change the way metadata is extracted for an existing channel, you can overwrite the function get_metadata. The return value of this method will be passed to the UserMessage.

Here are all the attributes of UserMessage:

class rasa.core.channels.UserMessage(text=None, output_channel=None, sender_id=None, parse_data=None, input_channel=None, message_id=None, metadata=None)

Represents an incoming message.

Includes the channel the responses should be sent to.

__init__(text=None, output_channel=None, sender_id=None, parse_data=None, input_channel=None, message_id=None, metadata=None)

Creates a UserMessage object.

  • text – the message text content.

  • output_channel – the output channel which should be used to send bot responses back to the user.

  • sender_id – the message owner ID.

  • parse_data – rasa data about the message.

  • input_channel – the name of the channel which received this message.

  • message_id – ID of the message.

  • metadata – additional metadata for this message.

In your implementation of the receive endpoint, you need to make sure to call on_new_message(UserMessage(text, output, sender_id)). This will tell Rasa Core to handle this user message. The output is an output channel implementing the OutputChannel class. You can either implement the methods for your particular chat channel (e.g. there are methods to send text and images) or you can use the CollectingOutputChannel to collect the bot responses Core creates while the bot is processing your messages and return them as part of your endpoint response. This is the way the RestInput channel is implemented. For examples on how to create and use your own output channel, take a look at the implementations of the other output channels, e.g. the SlackBot in rasa.core.channels.slack.

To use a custom channel, you need to supply a credentials configuration file credentials.yml with the command line argument --credentials. This credentials file has to contain the module path of your custom channel and any required configuration parameters. For example, this could look like:

  username: "user_name"
  another_parameter: "some value"

Here is an example implementation for an input channel that receives the messages, hands them over to Rasa Core, collects the bot utterances, and returns these bot utterances as the json response to the webhook call that posted the message to the channel:

class RestInput(InputChannel):
    """A custom http input channel.

    This implementation is the basis for a custom implementation of a chat
    frontend. You can customize this to send messages to Rasa Core and
    retrieve responses from the agent."""

    def name(cls) -> Text:
        return "rest"

    async def on_message_wrapper(
        on_new_message: Callable[[UserMessage], Awaitable[Any]],
        text: Text,
        queue: Queue,
        sender_id: Text,
        input_channel: Text,
        metadata: Optional[Dict[Text, Any]],
    ) -> None:
        collector = QueueOutputChannel(queue)

        message = UserMessage(
            text, collector, sender_id, input_channel=input_channel, metadata=metadata
        await on_new_message(message)

        await queue.put("DONE")  # pytype: disable=bad-return-type

    async def _extract_sender(self, req: Request) -> Optional[Text]:
        return req.json.get("sender", None)

    # noinspection PyMethodMayBeStatic
    def _extract_message(self, req: Request) -> Optional[Text]:
        return req.json.get("message", None)

    def _extract_input_channel(self, req: Request) -> Text:
        return req.json.get("input_channel") or self.name()

    def stream_response(
        on_new_message: Callable[[UserMessage], Awaitable[None]],
        text: Text,
        sender_id: Text,
        input_channel: Text,
        metadata: Optional[Dict[Text, Any]],
    ) -> Callable[[Any], Awaitable[None]]:
        async def stream(resp: Any) -> None:
            q = Queue()
            task = asyncio.ensure_future(
                    on_new_message, text, q, sender_id, input_channel, metadata
            result = None  # declare variable up front to avoid pytype error
            while True:
                result = await q.get()
                if result == "DONE":
                    await resp.write(json.dumps(result) + "\n")
            await task

        return stream  # pytype: disable=bad-return-type

    def blueprint(
        self, on_new_message: Callable[[UserMessage], Awaitable[None]]
    ) -> Blueprint:
        custom_webhook = Blueprint(

        # noinspection PyUnusedLocal
        @custom_webhook.route("/", methods=["GET"])
        async def health(request: Request) -> HTTPResponse:
            return response.json({"status": "ok"})

        @custom_webhook.route("/webhook", methods=["POST"])
        async def receive(request: Request) -> HTTPResponse:
            sender_id = await self._extract_sender(request)
            text = self._extract_message(request)
            should_use_stream = rasa.utils.endpoints.bool_arg(
                request, "stream", default=False
            input_channel = self._extract_input_channel(request)
            metadata = self.get_metadata(request)

            if should_use_stream:
                return response.stream(
                        on_new_message, text, sender_id, input_channel, metadata
                collector = CollectingOutputChannel()
                # noinspection PyBroadException
                    await on_new_message(
                except CancelledError:
                        "Message handling timed out for "
                        "user message '{}'.".format(text)
                except Exception:
                        "An exception occured while handling "
                        "user message '{}'.".format(text)
                return response.json(collector.messages)

        return custom_webhook