Hi, I saw a couple of other similar questions to mine in the community db but no answers. Hoping someone can shed some light.
I have a client app that’s using the node js library to authenticate a user. It receives and stores the resulting id_token
. Our Auth0 application is set up to use RS256. The client sends our python service the id_token
in the Authorization
header. The service is calling jwt.decode
to verify the id_token
, however it appears to be failing on the at_hash
which is included in the claims.
If I specify the access_token
argument to jwt.decode
, hard-coding it to the value I manually stole from the client after authing, it all works perfectly. However, that means that the client now has to send the service both the id_token
and the access_token
. My (admittedly hazy) understanding is that it’s using the access_token
to sign the id_token
, so sending them both across the wire seems counter-productive.
Is that accurate? How else can I validate the at_hash
successfully in a BE service, from a token created by a node app? Is it even necessary to verify the at_hash
?
Thanks much,
EH