This thread was very helpful to get started with a human hand-off in RASA, but there’s still some way to go before we have a working example. In this particular thread, code is provided that will allow us to send a message to a Slack channel when a user requests to talk to an agent. It is not discussed how the agent can forward messages back to the user and how the user can continue to engage with the agent until it is indicated that the bot can take over again. The main issue to solve is the following: how can we extend the code in order to send user messages to Slack and Slack messages to the user so long as the hand-off is required? The custom action provided in the thread looks as follows: import os from slack import WebClient from slack.errors import SlackApiError client_slack = WebClient(token=os.environ["SLACK_API_TOKEN"], run_async=True) SLACK_SUPPORT_CHANNEL = os.environ["SLACK_SUPPORT_CHANNEL"] class ActionHumanHandoff(Action): def name(self): return "action_human_handoff" async def run( self, dispatcher, tracker, domain, reason: Text = None, ): events = [] text = f"Help needed for user {tracker.sender_id}" try: response = await client_slack.chat_postMessage( channel=SLACK_SUPPORT_CHANNEL, text=text ) if response.get("ok"): dispatcher.utter_message(template="utter_handover_to_support") events.append(ConversationPaused()) except SlackApiError as e: dispatcher.utter_message( template="utter_technical_issue", extra_message=e.response["error"], ) return events

I think it’s not a good approach to use Rasa as the centralized messaging router. It’s not designed to play this role. Even if you implement the suggested solution, your tracker store will contain conversations that are unrelated to the domain of the bot. Instead, there should be a client UI that switches between Rasa and other services (Slack, another Rasa bot, etc). The action would send a message that the client UI recognizes as a transfer (via slot metadata) and pauses the conversation. The UI then connects to Slack or some other service.

@nchauras Hey Niraj, thanks for this post. It’s very helpful. Could you share the code for your policy? I’ve been trying to implement a policy that predicts an action when a slot is set but haven’t been having much luck

In your custom policy __init__ method, set the priority to highest int value (depends on your config and the priority for other policies). e.g. priority: int = <set to highest int value> and handoff_action_name: Text = "action_live_agent_chat". In predict_action_probabilities method, predict the handoff_action if tracker.get_slot('handoff_active') is True. Hope this helps!

@nchauras I put the priority in config.yml like so: ... - name: custom_policies.testpolicy priority: 6 the __init__.py file in the directory housing my custom action is empty. For some reason it seems that the policy is not actually running the action, though. Here’s my code in predict_action_probabilities: is_triggered = tracker.get_slot('my_slot') if is_triggered: logger.info('if statement ' + str(is_triggered) + ' ' + str(type(is_triggered))) return confidence_scores_for("action_my_action", 1.0, domain) else: logger.info('in else statement ' + str(is_triggered) + ' ' + str(type(is_triggered))) return self._default_predictions(domain) I also made sure to mount the file in docker-compose.override.yml so that shouldn’t be an issue

@stephens @nchauras @pcoutu Hi! Here we are trying to do human-handoff with Rasa bot. The hard part is retrieve the message from the human agent and send it to the user, but I do not want to do polling whitin an action to do that. Maybe something like this can be implemented: [Rasa handoff architecture] Can somebody help me with how to implement this? I’m thinking about rabbit or something like this but I’m not sure.

We recently added an enhancement to the chatroom Rasa widget here and we documented it’s use in the example helpdesk-assistant here . It is an example of what you show above without the use of a Router or Handoff DB. When the user requests a transfer, the Rasa response can include slots or metadata that is passed to the UI component. In the example implementation, it passes the URL to another Rasa chatbot but you could add your own metadata.

I ended up creating Webhook to push human messages to the bot using MS Teams API directly (as MS Teams is the front end for me). The messages from the user are pushed out to the human agent using rasa custom action & handoff policy and the human agent reply triggers webhook response directly to the front end. I followed this route to record the user messages as to why the human agent was requested and we do a weekly review to include those use-cases in the bot itself. This has helped us with the CDD tremendously.

Interesting! Does that mean that you review conversations containing both human chats and bot chats mixed together?

Hi @nchauras , Could you share the implementation code or repo to go through? Thanks.

Hi, were you able to implement this? Can you please share the steps you took

Forwarding messages to Slack and vice versa for human hand-off

Rasa Open Source

flore (Florencia Alonso) August 5, 2020, 6:08pm 16

Thank you, but I’m using whatsapp not a personal UI

Topic		Replies	Views
Handing off to Human agent Rasa Open Source	32	5291	July 14, 2021
Handing off to human on Slack Rasa Open Source	42	4580	December 2, 2021
Webchat with live operator [Deprecated] Rasa X Community Edition	3	853	July 10, 2023
Ideal approach for Human Handoff? Rasa Open Source	9	833	August 10, 2022
How to redirect conversations to human - rasa (socket, channel)) [Deprecated] Rasa X Community Edition	5	1769	February 28, 2022

Forwarding messages to Slack and vice versa for human hand-off

Related topics