dialogflow

问题 I tried to use Rich responses to show buttons, cards as responses from the webhook with dynamic content. agent.add(new Card({ title: <RICH UI>, imageUrl: '<IMAGE URL>', text: '<CARD TEXT>', buttonText: '<BUTTON>', buttonUrl: '<BUTTON URL>' }) ); agent.add(new Suggestion('<SUGGESTION 1>')); agent.add(new Suggestion('<SUGGESTION 2>')); It seems this is not supported for web-based Dialogflow agents. How can it be achieved like this guy is doing with his agent? I want to control the UI from

问题 Hi i'm trying to get device location of the user by asking permission.It works fine for simulator but when i tested with google mini device getting undefined for device location in webhook request.Below is the code const {Permission} = require('actions-on-google'); const {WebhookClient} = require('dialogflow-fulfillment'); const agent = new WebhookClient({ request: req, response: res }); conv.ask(new Permission({context:'To Locate You',permissions:'DEVICE_COARSE_LOCATION'})); function

问题 I am trying to get a simple webhook (written in PHP) to work with dialogflow/actions on google. I have a dialog flow intent labled "hello" which is linked to the "google assistant welcome" and dialog flow "welcome" events. it is set to enable webhook, and everything works correctly in the dialog flow test area. when i test it in google assistant, however, i get the following error: "MalformedResponse Failed to parse Dialogflow response into AppResponse." I have no clue what is wrong. Here is

问题 I'm befuddled as to how best fix this. In the training data, it looks like the Google agent parsed out the information correctly. But when I get it in my endpoint to fulfill the request, they do not match. There is no way for me to indicate that it made a mistake because according to the training information, it parsed everything correctly. Has anyone else encountered this, and if so, what was the solution? 回答1: Okay, I think I'm beginning to understand this a little. When this occurs, this

问题 I am trying to test my Dialogflow app on Actions on Google. I managed to release one alpha version without a problem but now when I try to release the second version it gives an error Testing instruction not found. And then nothing happens. I couldn't find this error anywhere, neither does it take to some dialog to fix it. What does this mean and how was I able to create one alpha version for the same project without error? 回答1: As we determined in the comments, you added account linking in

问题 Is it possible to pass some key value objects in the initial request to dialogflow which maybe needed in the webhook? 回答1: When you call /query , you can include key:value data in one of three ways: Include it in an object assigned to the originalRequest field. Include a context that has an object containing your key:value data in its parameters field. Trigger an event that has an object containing your key:value data in its data field. You can find this information in the details of the

问题 Is there a way to change how Dialogflow automatically calculates the date meant by a text identified as a pre-built entity like "sys.date" or "sys.date-period"? Or, is there a way to create your own entity for dates, and specify how it determines the specific day? Depending on the day of the week, texts like "on Friday" are actually interpreted as the closest Friday to come in the future, instead of the most recent Friday that has already passed. 回答1: If you are using @sys.date entity then

问题 After sending a MediaResponse and starting to stream audio, if the user says "pause", the Assistant pauses the playback, is there a way to get a callback to that event for tracking playback position purposes? 回答1: That is not currently supported. The only events you can handle with a callback is FINISHED or STATUS_UNSPECIFIED. 来源： https://stackoverflow.com/questions/52669444/is-there-a-way-to-get-media-callbacks-following-a-mediaresponse

问题 I am creating a Google action using Dialog flow with a webhook implemented as an Azure function written in c#. I have run into a problem that using .net core 2 that the response sent back from the Azure function is being chunked but that Dialog flow does not understand the chunked response. Though it was possible to switch off chunking with the full .net framework it seems harder to do with .net core 2. Has anyone found a solution to this problem? Thanks I searched online and is that I am not

问题 I am facing this issue in integrating Google SignIn flow. The problem I am facing is that after the user successfully sign in program control trigger actions_intent_SIGN_IN intent in the code. From where I can get user details but I am noticing that sometimes conv->user->profile doesn't have the profile information while other time it has. Also When it doesn't have profile information, next time when I invoke some other intent it gets the user token. This is the Payload I am getting on SignIn