M. Scurtescu | |
A. Backman | |
Amazon | |
J. Bradley | |
Yubico | |
April 24, 2018 |
OpenID RISC Profile of IETF Security Events 1.0
openid-risc-profile-1_0
This spec is a general profile for IETF Security Events and it defines:
This spec also directly profiles several IETF Security Events drafts:
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.
The RISC profile defines a structure called a Subject Identifier: a JSON [RFC7159] object containing a set of claims that collectively uniquely identify a subject, according to a simple schema called a Subject Identifier Type. Every Subject Identifier MUST contain a subject_type claim, whose value MUST be a name that uniquely identifies the Subject Identifier Type of the Subject Identifier. Any remaining claims within the Subject Identifier MUST be interpreted according to the definition of the Subject Identifier Type. A Subject Identifier MUST conform to the Subject Identifier Type identified by its subject_type claim, and MUST NOT contain any claims not defined by its Subject Identifier Type.
{ "iss": "https://idp.example.com/", "jti": "756E69717565206964656E746966696572", "iat": 1520364019, "aud": "636C69656E745F6964", "events": { "https://schemas.openid.net/secevent/risc/event-type/account-\ enabled": { "subject": { "subject_type": "email", "email": "foo@example.com" } } } }
A Subject Identifier that identifies a subject by email address.
Figure 1: Example: SET Containing a RISC Event with an Email Subject Identifier
{ "iss": "https://idp.example.com/", "jti": "756E69717565206964656E746966696572", "iat": 1520364019, "aud": "636C69656E745F6964", "events": { "https://schemas.openid.net/secevent/risc/event-type/account-\ enabled": { "subject": { "subject_type": "iss_sub", "iss": "https://issuer.example.com/", "sub": "abc1234" } } } }
A Subject Identifier that identifies a subject by an identifier provided by an issuer.
Figure 2: Example: SET Containing a RISC Event with an Issuer and Subject Subject Identifier
A Subject Identifier Type is a light-weight schema that describes a set of claims that uniquely identifies a subject. Every Subject Identifier Type MUST have a unique name registered in the IANA "Security Event Subject Identifier Types" registry established by Section 6.1. A Subject Identifier Type MAY contain more claims than are strictly necessary to uniquely identify a subject, however a Subject Identifier MUST contain all of the claims required by its type (See Privacy Considerations).
The following Subject Identifier Types are registered in the IANA "Security Event Subject Identifier Types" registry established by Section 6.1.
The Email Subject Identifier Type describes a subject by email address. Subject Identifiers of this type MUST contain an email claim whose value is a string containing the email address of the subject. The email claim MUST NOT be null or empty. The Email Subject Identifier Type is identified by the name email.
The Phone Number Subject Identifier Type describes a subject by telephone number. Subject Identifiers of this type MUST contain a phone claim whose value is a string containing the full telephone number of the subject, including international dialing prefix, formatted according to E.164 [E164]. The phone claim MUST NOT be null or empty. The Phone Number Subject Identifier Type is identified by the name phone.
The Issuer and Subject Subject Identifier Type describes a subject by an issuer and a subject. Subject Identifiers of this type MUST contain an iss claim whose value identifies the issuer, and a sub claim whose value identifies the subject with respect to the issuer. These claims MUST follow the formats of the iss claim and sub claim defined by [RFC7519], respectively. Both the iss claim and the sub claim MUST NOT be null or empty. The Issuer and Subject Subject Identifier Type is identified by the name iss_sub.
The ID Token Claims Subject Identifier Type describes a subject by a subset of the claims from an ID token. Subject Identifiers of this type MUST contain at least one of the following claims:
If the Subject Identifier contains a sub claim, it MUST also contain an iss claim, as defined in [RFC7519]. The ID Token Claims Subject Identifier Type is identified by the name id_token_claims.
This section defines a mechanism for Receivers to obtain Transmitter configuration information.
Transmitters have metadata describing their configuration:
TODO: consider adding a IANA Registry for metadata, similar to Section 7.1.1 of [OAUTH-DISCOVERY]. This would allow other specs to add to the metadata.
Using the Issuer as documented by the Transmitter, the Transmitter Configuration Information can be retrieved.
Transmitters supporting Discovery MUST make a JSON document available at the path formed by inserting the string /.well-known/risc-configuration into the Issuer between the host component and the path component, if any. The syntax and semantics of .well-known are defined in [RFC5785]. risc-configuration MUST point to a JSON document compliant with this specification and MUST be returned using the application/json content type.
A Transmitter Configuration Document MUST be queried using an HTTP GET request at the previously specified path.
GET /.well-known/risc-configuration HTTP/1.1 Host: tr.example.com
Figure 3: Example: Transmitter Configuration Request (without path)
The Receiver would make the following request to the Issuer https://tr.example.com to obtain its Configuration information, since the Issuer contains no path component:
GET /.well-known/risc-configuration/issuer1 HTTP/1.1 Host: tr.example.com
Figure 4: Example: Transmitter Configuration Request (with path)
If the Issuer value contains a path component, any terminating / MUST be removed before inserting /.well-known/risc-configuration between the host component and the path component. The Receiver would make the following request to the Issuer https://tr.example.com/issuer1 to obtain its Configuration information, since the Issuer contains a path component:
Using path components enables supporting multiple issuers per host. This is required in some multi-tenant hosting configurations. This use of .well-known is for supporting multiple issuers per host; unlike its use in [RFC5785], it does not provide general information about the host.
The response is a set of Claims about the Transmitter's configuration, including all necessary endpoints and public key location information. A successful response MUST use the 200 OK HTTP status code and return a JSON object using the application/json content type that contains a set of Claims as its members that are a subset of the Metadata values defined in Section 3.1. Other Claims MAY also be returned.
Claims that return multiple values are represented as JSON arrays. Claims with zero elements MUST be omitted from the response.
An error response uses the applicable HTTP status code value.
HTTP/1.1 200 OK Content-Type: application/json { "issuer": "https://tr.example.com", "jwks_uri": "https://tr.example.com/jwks.json", "delivery_methods_supported": [ "https://schemas.openid.net/secevent/risc/delivery-method/push", "https://schemas.openid.net/secevent/risc/delivery-method/poll"], "configuration_endpoint": "https://tr.example.com/risc/mgmt/stream", "status_endpoint": "https://tr.example.com/risc/mgmt/status", "add_subject_endpoint": "https://tr.example.com/risc/mgmt/subject:add", "remove_subject_endpoint": "https://tr.example.com/risc/mgmt/subject:remove", "verification_endpoint": "https://tr.example.com/risc/mgmt/verification" }
Figure 5: Example: Transmitter Configuration Response
If any of the validation procedures defined in this specification fail, any operations requiring the information that failed to correctly validate MUST be aborted and the information that failed to validate MUST NOT be used.
The issuer value returned MUST be identical to the Issuer URL that was directly used to retrieve the configuration information. This MUST also be identical to the iss Claim value in Security Event Tokens issued from this Transmitter.
This section defines an HTTP API to be implemented by Event Transmitters and that can be used by Event Receivers to query and update the Event Stream configuration and status, to add and remove subjects and to trigger verification.
This section is based on Management API for SET Event Streams [MGMTAPI].
+------------+ +------------+ | | Stream Config | | | Event <----------------+ Event | | Stream | | Receiver | | Management | Stream Status | | | API <----------------+ | | | | | | | Add Subject | | | <----------------+ | | | | | | | Remove Subject | | | <----------------+ | | | | | | | Verification | | | <----------------+ | | | | | +------------+ +------------+
Figure 6: Event Stream Management API
It is OPTIONAL for Transmitters to implement a Management API, but it is RECOMMENDED that they implement it, especially the endpoints for querying the Stream Status and for triggering Verification.
Event Receivers manage how they receive events, and the subjects about which they want to receive events over an Event Stream by making HTTP requests to endpoints in the Event Stream Management API.
The Event Stream Management API is implemented by the Event Transmitter and consists of the following endpoints:
An Event Transmitter MAY use the same URLs as endpoints for multiple streams, provided that the Event Transmitter has some mechanism through which they can identify the applicable Event Stream for any given request, e.g. from authentication credentials. The definition of such mechanisms is outside the scope of this specification.
An Event Receiver checks the current status of an event stream by making an HTTP GET request to the stream’s Status Endpoint. On receiving a valid request the Event Transmitter responds with a 200 OK response containing a JSON [RFC7159] object with a single attribute status, whose string value MUST have one of the following values:
The following is a non-normative example request to check an event stream’s status:
GET /set/status HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo=
Figure 7: Example: Check Stream Status Request
The following is a non-normative example response:
HTTP/1.1 200 OK Content-Type: application/json; charset=UTF-8 Cache-Control: no-store Pragma: no-cache { "status": "enabled" }
Figure 8: Example: Check Stream Status Response
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
401 | if authorization failed or it is missing |
403 | if the Event Receiver is not allowed to read the stream status |
404 | if there is no Event Stream configured for this Event Receiver |
An Event Receiver updates the current status of a stream by making an HTTP POST request to the Status Endpoint. The POST body contains a JSON [RFC7159] representation of the updated status. On receiving a valid request the Event Transmitter responds with a 200 OK response containing a JSON [RFC7159] representation of the updated stream status in the body.
The following is a non-normative example request to update an Event Stream’s status:
POST /set/status HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo= { "status": "paused", }
Figure 9: Example: Update Stream Status Request
The following is a non-normative example response:
HTTP/1.1 200 OK Content-Type: application/json; charset=UTF-8 Cache-Control: no-store Pragma: no-cache { "status": "paused" }
Figure 10: Example: Update Stream Status Response
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
400 | if the request body cannot be parsed or if the request is otherwise invalid |
401 | if authorization failed or it is missing |
403 | if the Event Receiver is not allowed to update the stream status |
404 | if there is no Event Stream configured for this Event Receiver |
An Event Stream’s configuration is represented as a JSON [RFC7159] object with the following properties:
TODO: consider adding a IANA Registry for stream configuration metadata, similar to Section 7.1.1 of [OAUTH-DISCOVERY]. This would allow other specs to add to the stream configuration.
An Event Receiver gets the current configuration of a stream by making an HTTP GET request to the Configuration Endpoint. On receiving a valid request the Event Transmitter responds with a 200 OK response containing a JSON [RFC7159] representation of the stream’s configuration in the body.
The following is a non-normative example request to read an Event Stream’s configuration:
GET /set/stream HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo=
Figure 11: Example: Read Stream Configuration Request
The following is a non-normative example response:
HTTP/1.1 200 OK Content-Type: application/json; charset=UTF-8 Cache-Control: no-store Pragma: no-cache { "iss": "https://tr.example.com", "aud": [ "http://receiver.example.com/web", "http://receiver.example.com/mobile" ], "delivery": { "delivery_method": "https://schemas.openid.net/secevent/risc/delivery-method/push", "url": "https://receiver.example.com/events" }, "events_supported": [ "urn:example:secevent:events:type_1", "urn:example:secevent:events:type_2", "urn:example:secevent:events:type_3" ], "events_requested": [ "urn:example:secevent:events:type_2", "urn:example:secevent:events:type_3", "urn:example:secevent:events:type_4" ], "events_delivered": [ "urn:example:secevent:events:type_2", "urn:example:secevent:events:type_3" ] }
Figure 12: Example: Read Stream Configuration Response
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
401 | if authorization failed or it is missing |
403 | if the Event Receiver is not allowed to read the stream configuration |
404 | if there is no Event Stream configured for this Event Receiver |
An Event Receiver updates the current configuration of a stream by making an HTTP POST request to the Configuration Endpoint. The POST body contains a JSON [RFC7159] representation of the updated configuration. On receiving a valid request the Event Transmitter responds with a 200 OK response containing a JSON [RFC7159] representation of the updated stream configuration in the body.
The full set of editable properties must be present in the POST body, not only the ones that are specifically intended to be changed. Missing properties SHOULD be interpreted as requested to be deleted. Event Receivers should read the configuration first, modify the JSON [RFC7159] representation, then make an update request.
Properties that cannot be updated MAY be present, but they MUST match the expected value.
The following is a non-normative example request to update an Event Stream’s configuration:
POST /set/stream HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo= { "iss": "https://tr.example.com", "aud": [ "http://receiver.example.com/web", "http://receiver.example.com/mobile" ], "delivery": { "delivery_method": "https://schemas.openid.net/secevent/risc/delivery-method/push", "url": "https://receiver.example.com/events" }, "events_requested": [ "urn:example:secevent:events:type_2", "urn:example:secevent:events:type_3", "urn:example:secevent:events:type_4" ] }
Figure 13: Example: Update Stream Configuration Request
The following is a non-normative example response:
HTTP/1.1 200 OK Content-Type: application/json; charset=UTF-8 Cache-Control: no-store Pragma: no-cache { "iss": "https://tr.example.com", "aud": [ "http://receiver.example.com/web", "http://receiver.example.com/mobile" ], "delivery": { "delivery_method": "https://schemas.openid.net/secevent/risc/delivery-method/push", "url": "https://receiver.example.com/events" }, "events_supported": [ "urn:example:secevent:events:type_1", "urn:example:secevent:events:type_2", "urn:example:secevent:events:type_3" ], "events_requested": [ "urn:example:secevent:events:type_2", "urn:example:secevent:events:type_3", "urn:example:secevent:events:type_4" ], "events_delivered": [ "urn:example:secevent:events:type_2", "urn:example:secevent:events:type_3" ] }
Figure 14: Example: Update Stream Configuration Response
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
400 | if the request body cannot be parsed or if the request is otherwise invalid |
401 | if authorization failed or it is missing |
403 | if the Event Receiver is not allowed to update the stream configuration |
404 | if there is no Event Stream configured for this Event Receiver |
An Event Receiver removes the configuration of a stream by making an HTTP DELETE request to the Configuration Endpoint. On receiving a request the Event Transmitter responds with a 200 OK response if the configuration was successfully removed.
The following is a non-normative example request to remove an Event Stream’s configuration:
DELETE /set/stream HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo=
Figure 15: Example: Remove Stream Configuration Request
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
401 | if authorization failed or it is missing |
403 | if the Event Receiver is not allowed to update the stream configuration |
An Event Receiver can indicate to an Event Transmitter whether or not the receiver wants to receive events about a particular subject by “adding” or “removing” that subject to the Event Stream, respectively.
To add a subject to an Event Stream, the Event Receiver makes an HTTP POST request to the Add Subject Endpoint, containing in the body a JSON object the following claims:
On a successful response, the Event Transmitter responds with an empty 200 OK response. The Event Transmitter MAY choose to silently ignore the request, for example if the subject has previously indicated to the transmitter that they do not want events to be transmitted to the Event Receiver. In this case, the transmitter MAY return an empty 200 OK response or an appropriate error code. See Security Considerations [management-sec].
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
400 | if the request body cannot be parsed or if the request is otherwise invalid |
401 | if authorization failed or it is missing |
403 | if the Event Receiver is not allowed to add this particular subject, or not allowed to add in general |
404 | if the subject is not recognized by the Event Transmitter, the Event Transmitter may chose to stay silent in this case and respond with 200 |
429 | if the Event Receiver is sending too many requests in a gvien amount of time |
The following is a non-normative example request to add a subject to a stream, where the subject is identified by an Email Subject Identifier.
POST /set/subjects:add HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo= { "subject": { "subject_type": "email", "email": "example.user@example.com" }, "verified": true }
Figure 16: Example: Add Subject Request
The following is a non-normative example response to a successful request:
HTTP/1.1 200 OK Server: transmitter.example.com Cache-Control: no-store Pragma: no-cache
Figure 17: Example: Add Subject Response
To remove a subject from an Event Stream, the Event Receiver makes an HTTP POST request to the Remove Subject Endpoint, containing in the body a JSON object with the following claims:
On a successful response, the Event Transmitter responds with a 204 No Content response.
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
400 | if the request body cannot be parsed or if the request is otherwise invalid |
401 | if authorization failed or it is missing |
403 | if the Event Receiver is not allowed to remove this particular subject, or not allowed to remove in general |
404 | if the subject is not recognized by the Event Transmitter, the Event Transmitter may chose to stay silent in this case and respond with 204 |
429 | if the Event Receiver is sending too many requests in a gvien amount of time |
The following is a non-normative example request where the subject is identified by a Phone Number Subject Identifier:
POST /set/subjects:remove HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo= { "subject": { "subject_type": "phone", "phone_number": "+1 206 555 0123" } }
Figure 18: Example: Remove Subject Request
The following is a non-normative example response to a successful request:
HTTP/1.1 204 No Content Server: transmitter.example.com Cache-Control: no-store Pragma: no-cache
Figure 19: Example: Remove Subject Response
In some cases, the frequency of event transmission on an Event Stream will be very low, making it difficult for an Event Receiver to tell the difference between expected behavior and event transmission failure due to a misconfigured stream. Event Receivers can request that a verification event be transmitted over the Event Stream, allowing the receiver to confirm that the stream is configured correctly upon successful receipt of the event. The acknowledgment of a Verification Event also confirms to the Event Transmitter that end-to-end delivery is working, including signature verification and encryption.
An Event Transmitter MAY send a Verification Event at any time, even if one was not requested by the Event Receiver.
The Verification Event is a standard SET with the following attributes:
Upon receiving a Verification Event, the Event Receiver SHALL parse the SET and validate its claims. In particular, the Event Receiver SHALL confirm that the value for state is as expected. If the value of state does not match, an error response of setData SHOULD be returned (see Section 2.3 of [DELIVERYPUSH] or [DELIVERYPOLL]).
In many cases, Event Transmitters MAY disable or suspend an Event Stream that fails to successfully verify based on the acknowledgement or lack of acknowledgement by the Event Receiver.
To request that a verification event be sent over an Event Stream, the Event Receiver makes an HTTP POST request to the Verification Endpoint, with a JSON [RFC7159] object containing the parameters of the verification request, if any. On a successful request, the event transmitter responds with an empty 204 No Content response.
Verification requests have the following properties:
A successful response from a POST to the Verification Endpoint does not indicate that the verification event was transmitted successfully, only that the Event Transmitter has transmitted the event or will do so at some point in the future. Event Transmitters MAY transmit the event via an asynchronous process, and SHOULD publish an SLA for verification event transmission times. Event Receivers MUST NOT depend on the verification event being transmitted synchronously or in any particular order relative to the current queue of events.
Errors are signaled with HTTP status codes as follows:
Code | Description |
---|---|
400 | if the request body cannot be parsed or if the request is otherwise invalid |
401 | if authorization failed or it is missing |
429 | if the Event Receiver is sending too many requests in a gvien amount of time; see related min_verification_interval in Section 4.1.2 |
The following is a non-normative example request to trigger a verification event:
POST /set/verify HTTP/1.1 Host: transmitter.example.com Authorization: Bearer eyJ0b2tlbiI6ImV4YW1wbGUifQo= Content-Type: application/json; charset=UTF-8 { "state": "VGhpcyBpcyBhbiBleGFtcGxlIHN0YXRlIHZhbHVlLgo=" }
Figure 20: Example: Trigger Verification Request
The following is a non-normative example response to a successful request:
HTTP/1.1 204 No Content Server: transmitter.example.com Cache-Control: no-store Pragma: no-cache
Figure 21: Example: Trigger Verification Response
And the following is a non-normative example of a verification event sent to the Event Receiver as a result of the above request:
{ "jti": "123456", "iss": "https://transmitter.example.com", "aud": "receiver.example.com", "iat": "1493856000", "events": [ "https://schemas.openid.net/secevent/risc/event-type/verification":{ "state": "VGhpcyBpcyBhbiBleGFtcGxlIHN0YXRlIHZhbHVlLgo=" } ] }
Figure 22: Example: Verification SET
HTTP API calls from a Receiver to a Transmitter SHOULD be authorized by providing an OAuth 2 Access Token as defined by [RFC6750].
The receiver may obtain an access token using the Client Credential Grant [CLIENTCRED], or any other method suitable for the Receiver and the Transmitter.
Other authorization methods MAY be used if the Transmitter and the Receiver have agreement on the method.
It may be possible for an Event Transmitter to leak information about subjects through their responses to add subject requests. A 404 response may indicate to the Event Receiver that the subject does not exist, which may inadvertantly reveal information about the subject (e.g. that a particular individual does or does not use the Event Transmitter’s service).
Event Transmitters SHOULD carefully evaluate the conditions under which they will return error responses to add subject requests. Event Transmitters MAY return a 204 response even if they will not actually send any events related to the subject, and Event Receivers MUST NOT assume that a 204 response means that they will receive events related to the subject.
SETs may contain personally identifiable information (PII) or other non-public information about the event transmitter, the subject (of an event in the SET), or the relationship between the two. It is important for Event Transmitters to understand what information they are revealing to Event Receivers when transmitting events to them, lest the event stream become a vector for unauthorized access to private information.
Event Transmitters SHOULD interpret add subject requests as statements of interest in a subject by an Event Receiver, and ARE NOT obligated to transmit events related to every subject an Event Receiver adds to the stream. Event Transmitters MAY choose to transmit some, all, or no events related to any given subject and SHOULD validate that they are permitted to share the information contained within an event with the Event Receiver before transmitting the event. The mechanisms by which such validation is performed are outside the scope of this specification.
A malicious party may find it advantageous to remove a particular subject from a stream, in order to reduce the Event Receiver’s ability to detect malicious activity related to the subject, inconvenience the subject, or for other reasons. Consequently it may be in the best interests of the subject for the Event Transmitter to continue to send events related to the subject for some time after the subject has been removed from a stream.
Event Transmitters MAY continue sending events related to a subject for some amount of time after that subject has been removed from the stream. Event Receivers MUST tolerate receiving events for subjects that have been removed from the stream, and MUST NOT report these events as errors to the Event Transmitter.
This section is a profile of the following IETF SecEvent specifications:
The RISC use cases that set the requirements are described in Security Events RISC Use Cases [USECASES].
This section provides RISC profiling specifications for the "Security Event Token (SET)" [SET] spec.
The signature key can be obtained through jwks_uri, see Section 3.
The subject of a RISC event is identified by the subject claim within the event payload, whose value is a Subject Identifier. The subject claim is REQUIRED for all RISC events. The JWT sub claim MUST NOT be present in any SET containing a RISC event.
{ "iss": "https://idp.example.com/", "jti": "756E69717565206964656E746966696572", "iat": 1520364019, "aud": "636C69656E745F6964", "events": { "https://schemas.openid.net/secevent/risc/event-type/account-\ disabled": { "subject": { "subject_type": "phone", "phone_number": "+1 206 555 0123" }, "reason": "hijacking", "cause-time": 1508012752 } } }
Figure 23: Example: SET Containing a RISC Event with a Phone Number Subject
RISC events MUST use explicit typing as defined in Section 2.3 of [SET].
{ "typ":"secevent+jwt", "alg":"HS256" }
Figure 24: Explicitly Typed JOSE Header
The purpose is defense against confusion with other JWTs, as described in Sections 4.5, 4.6 and 4.7 of [SET]. While current Id Token [IDTOKEN] validators may not be using the typ header parameter, by requiring it for RISC SETs a distinct value is guaranteed for future validators.
The exp claim MUST NOT be used in RISC SETs.
The purpose is defense in depth against confusion with other JWTs, as described in Sections 4.5 and 4.6 of [SET].
The aud claim can be a single value or an array. Each value SHOULD be the OAuth 2 client id. Other values that uniquely identifies the Receiver to the Transmitter MAY be used, if the two parties have agreement on the format.
More than one values can be present if the corresponding Receivers are known to the Transmitter to be the same entity, for example a web client and I mobile client of the same application. All the Receivers in this case MUST use the exact same delivery method.
If multiple Receivers have the exact same delivery configuration but the Transmitter does not know if they belong to the same entity then the Transmitter SHOULD issue distinct SETs for each Receiver and deliver them separately. In this case the multiple Receivers might use the same service to process SETs, and this service might reroute SETs to respective Receivers, an aud claim with multiple Receivers would lead to unintended data disclosure.
{ "jti": "123456", "iss": "https://transmitter.example.com", "aud": ["receiver.example.com/web", "receiver.example.com/mobile"], "iat": "1493856000", "events": [ "https://schemas.openid.net/secevent/risc/event-type/verification":{ "state": "VGhpcyBpcyBhbiBleGFtcGxlIHN0YXRlIHZhbHVlLgo=" } ] }
Figure 25: Example: SET with array 'aud' claim
The events claim SHOULD contain only one event. Multiple event type URIs are permitted only if they are alternative URIs defining the exact same event type.
Of particular concern is the possibility that SETs are confused for other kinds of JWTs. The Security Considerations sesction of [SET] has several sub-sections on this subject. The RISC Profile is asking for further restrioctons:
This section provides RISC profiling specifications for the [DELIVERYPUSH] and [DELIVERYPOLL] specs.
Each delivery method is identified by a URI, specified below by the method metadata.
This section provides RISC profiling specifications for the [DELIVERYPUSH] spec.
This section provides RISC profiling specifications for the [DELIVERYPOLL] spec.
This document defines Subject Identifier Types, for which IANA is asked to create and maintain a new registry titled "Security Event Subject Identifier Types". Initial values for the Security Event Subject Identifier Types registry are given in Section 6.1.2. Future assignments are to be made through the Expert Review registration policy [BCP26] and shall follow the template presented in Section 6.1.1.
The Expert Reviewer is expected to review the documentation referenced in a registration request to verify its completeness. The Expert Reviewer must base their decision to accept or reject the request on a fair and impartial assessment of the request. If the Expert Reviewer has a conflict of interest, such as being an author of a defining document referenced by the request, they must recuse themselves from the approval process for that request. In the case where a request is rejected, the Expert Reviewer should provide the requesting party with a written statement expressing the reason for rejection, and be prepared to cite any sources of information that went into that decision.
Subject Identifier Types need not be generally applicable and may be highly specific to a particular domain; it is expected that types may be registered for niche or industry-specific use cases. The Expert Reviewer should focus on whether the type is thoroughly documented, and whether its registration will promote or harm interoperability. In most cases, the Expert Reviewer should not approve a request if the registration would contribute to confusion, or amount to a synonym for an existing type.
TODO: follow steps in Section 5.1 of [RFC5785] to send a registration request.
This specification registers the well-known URI defined in Section 3 in the IANA Well-Known URI registry defined in [RFC5785].
Event issuers and recipients SHOULD take precautions to ensure that they do not leak information about subjects via Subject Identifiers, and choose appropriate Subject Identifier Types accordingly. Parties SHOULD NOT identify a subject using a given Subject Identifier Type if doing so will allow the recipient to correlate different claims about the subject that they are not known to already have knowledge of. Issuers and recipients SHOULD always use the same Subject Identifier Type and the same claim values to identify a given subject when communicating with a given party in order to reduce the possibility of information leakage.
Transmitter Configuration Discovery [discovery] is based on both OpenID Connect Discovery 1.0 [OIDC-DISCOVERY] and OAuth 2.0 Authorization Server Metadata [OAUTH-DISCOVERY].