Home Community Blog Buy Now
Blog

Misty Community Forum

Concierge Skill Authentication Error

Hi,

I am trying to setup the concierge skill on misty.

I get the following error when calling DialogFlow

Request had invalid authentication credentials. Ex…ogle.com/identity/sign-in/web/devconsole-project.", status: “UNAUTHENTICATED”

Below is some info. It is getting an access code. And verified it is sending it to the call to detect intent.

AUTH->CjxodG1sPjxoZWFkPgo8bWV0YSBodHRwLWVxdWl2PSJjb250ZW50LXR5cGUiIGNvbnRlbnQ9InRleHQvaHRtbDtjaGFyc2V0PXV0Zi04Ij4KPHRpdGxlPjQwMyBGb3JiaWRkZW48L3RpdGxlPgo8L2hlYWQ+Cjxib2R5IHRleHQ9IzAwMDAwMCBiZ2NvbG9yPSNmZmZmZmY+CjxoMT5FcnJvcjogRm9yYmlkZGVuPC9oMT4KPGgyPllvdXIgY2xpZW50IGRvZXMgbm90IGhhdmUgcGVybWlzc2lvbiB0byBnZXQgVVJMIDxjb2RlPi9hdXRoVG9rZW48L2NvZGU+IGZyb20gdGhpcyBzZXJ2ZXIuPC9oMj4KPGgyPjwvaDI+CjwvYm9keT48L2h0bWw+Cg==
skillrunner.js:511 "‘conciergeBaseTemplate’ Debug => " “2020-07-26T00:29:38.1122364Z”
skillrunner.js:518 URL=>https://dialogflow.googleapis.com/v2/projects/newagent-rlfcsw/agent/sessions/123456:detectIntent
skillrunner.js:511 "‘conciergeBaseTemplate’ Debug => " “2020-07-26T00:29:38.5647378Z”
skillrunner.js:518 Response Parsed
skillrunner.js:511 "‘conciergeBaseTemplate’ Debug => " “2020-07-26T00:29:38.5647378Z”
skillrunner.js:515

It appears the return from

misty.SendExternalRequest("POST", misty.Get("cloudFunctionAuthTokenURL"), null, null, null, false, false, null, "application/json", "_UpdateAuthToken");

The data returned, there is no authToken as part of Data. It looks encrytped…

Returnd Value1->{“Status”:3,“Result”:{“Command”:null,“ResponseObject”:{“Data”:“CjxodG1sPjxoZWFkPgo8bWV0YSBodHRwLWVxdWl2PSJjb250ZW50LXR5cGUiIGNvbnRlbnQ9InRleHQvaHRtbDtjaGFyc2V0PXV0Zi04Ij4KPHRpdGxlPjQwMyBGb3JiaWRkZW48L3RpdGxlPgo8L2hlYWQ+Cjxib2R5IHRleHQ9IzAwMDAwMCBiZ2NvbG9yPSNmZmZmZmY+CjxoMT5FcnJvcjogRm9yYmlkZGVuPC9oMT4KPGgyPllvdXIgY2xpZW50IGRvZXMgbm90IGhhdmUgcGVybWlzc2lvbiB0byBnZXQgVVJMIDxjb2RlPi9hdXRoVG9rZW48L2NvZGU+IGZyb20gdGhpcyBzZXJ2ZXIuPC9oMj4KPGgyPjwvaDI+CjwvYm9keT48L2h0bWw+Cg==”,“ContentType”:“text/html”}}}

@wlaubach64,

This is Base64 encoded. Decoding it reads “Your client does not have permission to get URL /authToken from this server.”.

  1. Check this in the cloud function and that would let the skill pull the access token.

    If you would like to add login credentials; you could modify the skill and cloud function similar to the conversation skill you worked on earlier.

Note: The cloud function and the dialogue flow agent should both be under the same Google Cloud Project ID/Name.

Let me know if this works out !! :slight_smile:
CP [• •]

Thank you CP!!!

That fixed the problem…I really appreciate your help

Wes

1 Like