ValidatingTokens

Token Validation for a bearer token has multiple parts. The token is validated by checking that it is for the application, that it was issued by a trustable Security Token Server (STS), that the token's lifetime is in range, and that it was not tampered with. There can also be special validations. For instance, it is possible to validate that signing keys (when embedded in a token) are trusted and that the token is not being replayed. Finally, some protocols require specific validations.

Validators

The validation steps are captured into Validators, which are all in one source file: Microsoft.IdentityModel.Tokens/Validators.cs

The validators are the following:

Validator	Description
ValidateAudience	Ensures that the token is indeed for the application that validates the token (for me)
ValidateIssuer	Ensures that the token was issued by a STS I trust (from someone I TRUST)
ValidateIssuerSigningKey	Ensures the application validating the token trusts the key that was used to sign the token (this is a special case where the key is embedded in the token, usually this is not required)
ValidateLifetime	Ensures that the token is still (or already) valid. This is done by checking that the lifetime of the token (notbefore, expires) is in range
ValidateTokenReplay	Ensure the token is not replayed (this is a special case for some onetime use protocols)

Protocol specific validators

In addition to these validators, there are protocol specific validation rules. For example, OpenIdConnect requires the audience (‘aud’) claim to exist. See:Microsoft.IdentityModel.Protocols.OpenIdConnect/OpenIdConnectProtocolValidator.cs#L382