# Alexa Skill with RASA X

**URL:** <https://forum.rasa.com/t/alexa-skill-with-rasa-x/45839>\
**Category:** \[Deprecated\] Rasa X Community Edition\
**Created:** [July 16, 2021, 8:59pm UTC](https://forum.rasa.com/t/alexa-skill-with-rasa-x/45839 "2021-07-16T20:59:14Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Vin](https://avatars.discourse-cdn.com/v4/letter/v/a87d85/32.png) [@Vin](https://forum.rasa.com/u/Vin)\
**Post date:** [July 16, 2021, 8:59pm UTC](https://forum.rasa.com/t/alexa-skill-with-rasa-x/45839/1 "2021-07-16T20:59:14Z")

</div>

Hi everyone, I’m new to the Rasa community but I’m really enjoying the framework … have had a surprisingly good experience with Rasa open source but a bit more arduous one with Rasa X !!

I have been able to connect an Alexa skill with Rasa X (using docker-compose method on server), store and monitor the conversations on UI. But my deployment (at least for the custom Alexa channel) is incorrect since I see all conversations coming from different users invoking the Alexa skill under the same conversation on the Rasa X UI.

So how do I deploy my chatbot to handle multiple User invocations (specifically for the custom Alexa channel) at the same time, while creating separate conversations on the Rasa X UI ? Any help would greatly appreciated…Thanks !

---

<div class="post-metadata">

**Author:** ![Vin](https://avatars.discourse-cdn.com/v4/letter/v/a87d85/32.png) [@Vin](https://forum.rasa.com/u/Vin)\
**Post date:** [July 20, 2021, 8:46pm UTC](https://forum.rasa.com/t/alexa-skill-with-rasa-x/45839/2 "2021-07-20T20:46:13Z")

</div>

Any suggestions or ideas on how to create a new session/conversation tab for every user invocation from a custom channel ?

---

<div class="post-metadata">

**Author:** ![Vin](https://avatars.discourse-cdn.com/v4/letter/v/a87d85/32.png) [@Vin](https://forum.rasa.com/u/Vin)\
**Post date:** [July 21, 2021, 9:41am UTC](https://forum.rasa.com/t/alexa-skill-with-rasa-x/45839/3 "2021-07-21T09:41:32Z")

</div>

Posting the solution in case someone has a similar issue… in custom\_connector.py, while calling the Rasa open source add a “sender\_id” parameter.

In case of Alexa custom connector use the session-id values from the request

_sessionId = payload[“session”][“sessionId”]_

 

_await on\_new\_message(UserMessage(text, out, input\_channel=input\_channel, **sender\_id=sessionId** ))_
