Reading from http://en.wikipedia.org/wiki/SAML_2.0 , I am trying to understand the mechanism that guarantees the authenticity of the SAMLResponse sent in step 5 of the process.
As I understand it, SAML is designed in such a way that a service provider can trust an authentication provider for authentication without direct contact with it. However, this depends on how the user agent retransmits the response from the identity provider back to the service provider. It seems to me that a compromised agent can return a fake confirmed answer. How will the service provider confirm this without trying to contact the identity provider?
Thank you for your time.
source
share