Questions about App Validation and Bots Policy for a Notification-Only Bot in Microsoft Teams
Hello, I have a question regarding the app validation process and bot policies for notification-only bots in Microsoft Teams.
I am currently developing a Teams bot to notify users of updates in our web application. This bot is intended to be published in the Teams Store in the future. The main requirements of this bot are as follows:
- Notification-Only: No interactive communication with users.
- 1-on-1 Chat: Notifications are sent in a 1-on-1 chat format because the information depends on user permissions.
I created a Teams app manifest that meets these requirements. However, during the app validation process, the following error was encountered:
The bot is not responding
Upon reviewing the skipped test cases, it appears that the error occurred because the bot does not respond to three specific commands: Hi
, Hello
, and Help
.
Indeed, I have not implemented these commands because the bot I am developing is designed solely for notifications and does not require interaction with users. To ensure this, I have set isNotificationOnly
to true
in the manifest to disable interactive functionality.
Additionally, considering development costs and resource efficiency, I have decided not to implement endpoints using frameworks like Express or Restify. Instead, my bot sends proactive messages directly to the Bot Framework REST API, which I have successfully tested.
Given these circumstances, I would greatly appreciate any guidance on whether it is possible to avoid implementing the aforementioned commands.
To reiterate, implementing these commands would not align with the purpose of our bot and would increase development costs, operational burdens, and maintenance requirements, which we believe should be avoided.
If implementing these commands is mandatory, I would be grateful if you could share any references or examples for deploying the bot using AWS Lambda instead of Azure.
Additionally, I hope for potential future changes to this policy.
Thank you very much for your support and understanding.
Below is my Teams app manifest. Sensitive information has been replaced with hoge
.
{
"$schema": "https://developer.microsoft.com/en-us/json-schemas/teams/v1.19/MicrosoftTeams.schema.json",
"version": "1.0.0",
"manifestVersion": "1.19",
"id": "hoge",
"name": {
"short": "hoge",
"full": ""
},
"developer": {
"name": "hoge",
"mpnId": "hoge",
"websiteUrl": "hoge",
"privacyUrl": "hoge",
"termsOfUseUrl": "hoge"
},
"description": {
"short": "hoge",
"full": "hoge"
},
"icons": {
"outline": "outline.png",
"color": "color.png"
},
"accentColor": "#FFFFFF",
"staticTabs": [
{
"entityId": "conversations",
"scopes": [
"personal"
]
},
{
"entityId": "about",
"scopes": [
"personal"
]
}
],
"bots": [
{
"botId": "hoge",
"scopes": [
"personal"
],
"isNotificationOnly": true,
"supportsCalling": false,
"supportsVideo": false,
"supportsFiles": false
}
],
"validDomains": []
}
I am facing the same issue as the users in the following threads:
- https://learn.microsoft.com/en-us/answers/questions/2087531/clarification-on-response-requirements-for-one-way
- https://learn.microsoft.com/en-us/answers/questions/2087561/need-clarity-for-one-way-bot
If this is a bug on Microsoft's side, I sincerely hope it will be resolved promptly.