| Internet-Draft | json-proof-token | April 2023 |
| Miller, et al. | Expires 28 October 2023 | [Page] |
JSON Proof Token (JPT) is a compact, URL-safe, privacy-preserving representation of claims to be transferred between three parties. The claims in a JPT are encoded as base64url-encoded JSON objects that are used as the payloads of a JSON Web Proof (JWP) structure, enabling them to be digitally signed and selectively disclosed. JPTs also support reusability and unlinkability when using Zero-Knowledge Proofs (ZKPs).¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 28 October 2023.¶
Copyright (c) 2023 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
JSON Proof Token (JPT) is a compact claims representation format intended to be used in the same ways as a JSON Web Token (JWT), but with additional support for selective disclosure and unlinkability. JPTs encode claim values to be transmitted as payloads of a JSON Web Proof (JWP). JPTs are always represented using the JWP Compact Serialization. The corresponding claim names are not transmitted in the payloads and are stored in a separate structure that can be externalized and shared across multiple JPTs.¶
Editor's Note: This draft is still early and incomplete, there will be significant changes to the algorithms as currently defined here. Please do not use any of these definitions or examples for anything except personal experimentation and learning. Contributions and feedback are welcome at https://github.com/json-web-proofs/json-web-proofs.¶
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.¶
JWP defines a container binding together a protected header, one or more payloads, and a cryptographic proof. It does not define how claims are organized into payloads and what formats they are in. JPTs are intended to be as close to a JWT as possible, while also supporting the selective disclosure and unlinkability of JWPs.¶
The rationale behind the design for JSON Proof Tokens is important when considering how it is structured. These sections detail the underlying reasoning for the approach defined by JPTs.¶
Supporting unlinkability is perhaps the most challenging design constraint for JPTs. Even the smallest oversight can introduce a subtle vector for relying parties to collude and correlate one or more subjects across their usage.¶
The principal tools to prevent this are data minimization and uniformity. The data included in a JPT SHOULD be minimized to remove potential correlation points. The data SHOULD contain only values that are able to be selectively disclosed with consent or transformed by the proof algorithm when presented.¶
Any other data that is repeated across multiple JPTs is externalized so that it is uniform across every issuance. This includes preventing the usage of optional headers, dynamic mapping of claims to payloads, changes to how many payloads are included, and the ordering of the payloads.¶
While JWPs provide the underling structure for easily supporting selective disclosure, JPTs must go a step further to ensure that applications can effectively provide choice and consent on exactly what is being disclosed. Software using JWPs must know the mappings from payloads to claims. JPTs do not support disclosing claims that are intended to be private from the issuer to the relying party. All disclosed payloads MUST be mapped to claims and made accessible to the application.¶
JPTs are intended to be as close to a JWT as possible in order to provide the simplest transition for any JWT-based system to add support for a JPT.¶
Although there are some stark differences in the lifecycle of a JPT, from the application's perspective, the interface to a JPT can be made fairly similar: a JSON object containing a mix of required and optional claims with well-understood values.¶
The most significant divergence required by JPTs is that of supporting values that may be disclosed or may instead only be a proof about the value. Applications are required to interact with the JPT on a payload-by-payload basis instead of just verifying a JWT and then being able to interact with the JSON body directly.¶
In order to generate a variety of efficient ZKPs of knowledge, range, membership, or other predicates, it is essential that each individual payload is only a single claim value. This greatly simplifies the task of linking a derived proof of a given claim to the specific payload that was also signed by the issuer. While JPTs support claims that have complex object or array compound values, they also allow for simple claim values such as JSON strings, numbers, and booleans that can be used directly in generating predicate proofs.¶
It is suggested that the claim names used with JPTs come from those in the IANA JSON Web Token Claims Registry, when those fit the application's needs.¶
Using a JSON Proof Token requires combining information from two sources: the claim names and the payloads. The simplest solution is to list the claim names in an ordered array that aligns with the included payloads. This claims array can be conveniently included in the JWP Protected Header using the claims key.¶
When the claims array is stored in the header, any variations of it are disclosed to the verifier and can be used to correlate and link usages. Given the privacy design considerations around linkability it is recommended that the claims are defined external to an individual JPT and either referenced or known by the application context.¶
In order to facilitate this external definition of the claim names, an additional cid key is defined with a required digest value calculated as defined here. This cid can be used similar to a kid in order to ensure externally resolve and then verify that the correct list of claim names is being used when processing the payloads containing the claim values.¶
If there is an associated JWK containing the signing key information, the claims key is also registered there as a convenient location for the claim names.¶
All payloads are claim values and MUST be the base64url encoding of the UTF-8 representation of a JSON value.¶
The following is an example JWP Protected Header that includes a claims array:¶
{
"kid": "HjfcpyjuZQ-O8Ye2hQnNbT9RbbnrobptdnExR0DUjU8",
"alg": "BBS",
"claims": [
"iat",
"exp",
"family_name",
"given_name",
"email"
]
}
¶
Editor's Note: This section is significantly incomplete, use it only as an indicator of the intended direction.¶
Application resolves each claim as required when processing the JPT. Resolution can result in one of three things:
1. A disclosed JSON value
2. A custom proof method
3. A null value¶
See the JSON Web Proof draft appendix.¶
This document has no IANA actions.¶
[[ To be removed from the final specification ]]¶
-00¶