]> Fraunhofer SIT

Rheinstrasse 75 Darmstadt 64295 Germany henk.birkholz@ietf.contact

Bowball Technologies Ltd

United Kingdom jonathan@bowball-tech.com

Microsoft Research

21 Station Road Cambridge CB1 2FB UK antdl@microsoft.com

Security SCITT Internet-Draft This document describes a REST API that supports the normative requirements of the Supply Chain Integrity, Transparency, and Trust (SCITT) Architecture. About This Document Status information for this document may be found at . Discussion of this document takes place on the SCITT Working Group mailing list (), which is archived at . Subscribe at . Source for this draft and an issue tracker can be found at .

Introduction The Supply Chain Integrity, Transparency, and Trust (SCITT) Architecture defines the core objects, identifiers and workflows necessary to interact with a SCITT Transparency Service:

• Signed Statements
• Receipts
• Transparent Statements
• Registration Policies

SCITT Reference APIs (SCRAPI) defines HTTP resources for a Transparency Service using COSE (). The following resources MUST be implemented for conformance to this specification:

• Registration of Signed Statements
• Issuance and resolution of Receipts
• Discovery of Transparency Service Keys

Terminology 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 when, and only when, they appear in all capitals, as shown here. This specification uses the terms "Signed Statement", "Receipt", "Transparent Statement", "Artifact Repositories", "Transparency Service" and "Registration Policy" as defined in . This specification uses "payload" as defined in .

Resources All messages are sent as HTTP GET or POST requests. If the Transparency Service cannot process a client's request, it MUST return either:

1. an HTTP 3xx code, indicating to the client additional action they must take to complete the request, such as follow a redirection, or
2. an HTTP 4xx or 5xx status code, and the body MUST be a Concise Problem Details object (application/concise-problem-details+cbor) containing:

• title: A human-readable string identifying the error that prevented the Transparency Service from processing the request, ideally short and suitable for inclusion in log messages.
• detail: A human-readable string describing the error in more depth, ideally with sufficient detail enabling the error to be rectified.

SCRAPI is not a CoAP API, but Constrained Problem Details objects provide a useful encoding for problem details and avoid the need to mix CBOR and JSON in resource or client implementations. NOTE: Examples use '\' line wrapping per Examples of errors may include: Most error types are specific to the type of request and are defined in the respective subsections below. The one exception is the "malformed" error type, which indicates that the Transparency Service could not parse the client's request because it did not comply with this document: Clients SHOULD treat 500 and 503 HTTP status code responses as transient failures and MAY retry the same request without modification at a later date. Note that in the case of any error response, the Transparency Service MAY include a Retry-After header field per in order to request a minimum time for the client to wait before retrying the request. In the absence of this header field, this document does not specify a minimum. The following HTTP resources MUST be implemented to enable conformance to this specification.

Transparency Service Keys This resource is used to discover the public keys that can be used by relying parties to verify Receipts issued by the Transparency Service. The Transparency Service responds with a COSE Key Set, as defined in . Request: Response: The Transparency Service MAY stop returning at that resource the keys it no longer uses to issue Receipts, following a reasonable delay. A delay is considered reasonable if it is sufficient for relying parties to have obtained the key needed to verify any previously issued Receipt. Consistent with key management best practices described in (Section 5.3.4), retired keys SHOULD remain available for as long as any Receipts signed with them may still need to be verified.

Individual Transparency Service Key This resource is used to resolve a single public key, from a kid value contained in a Receipt previously issued by the Transparency Service. The Transparency Service responds with a single COSE Key, as defined in . Request: Response: The following expected error is defined and MUST be returned when the corresponding condition is encountered: Implementations MAY return other errors, so long as they are valid objects. If the kid values used by the service ({kid_value} in the request above) are not URL-safe, the resource MUST accept the base64url encoding of the kid value, without padding, in the URL instead. specifies Base64Url encoding as follows: specifies Base64url encoding as follows: "Base64 encoding using the URL- and filename-safe character set defined in Section 5 of RFC 4648 , with all trailing '=' characters omitted and without the inclusion of any line breaks, whitespace, or other additional characters. Note that the base64url encoding of the empty octet sequence is the empty string. (See Appendix C of for notes on implementing base64url encoding without padding.)" It is RECOMMENDED to use COSE Key Thumbprint, as defined in as the mechanism to assign a kid to Transparency Service keys.

Register Signed Statement This resource instructs a Transparency Service to register a Signed Statement on its log. Since log implementations may take many seconds or longer to reach finality, this API provides an asynchronous mode that returns a locator that can be used to check the registration's status asynchronously. The following is a non-normative example of an HTTP request to register a Signed Statement: Request: >, / Protected Header / {}, / Unprotected Header / / Payload, sha-256 digest of file stored at payload-location / h'935b5a91...e18a588a', h'269cd68f4211dffc...0dcb29c' / Signature / ]) ]]> A Transparency Service depends on the verification of the Signed Statement in the Registration Policy. The Registration Policy for the Transparency Service MUST be applied before any additional processing. The details of Registration Policies are out of scope for this document. Signed Statements MAY use detached payloads, as described in . When a Signed Statement is submitted with a detached payload, the Transparency Service still requires access to the payload content in order to verify the signature as part of applying the Registration Policy. The mechanism by which the payload is made available to the Transparency Service is implementation-specific. Response: One of the following:

Status 201 - Registration is successful If the Transparency Service is able to produce a Receipt within a reasonable time, it MAY return it directly. Along with the receipt the Transparency Service MAY return a locator in the HTTP response Location header, provided the locator is a valid URL. >, / unprotected / { / proofs / 396 : { / inclusion / -1 : [ <<[ / size / 9, / leaf / 8, / inclusion path / h'7558a95f...e02e35d6' ]>> ], }, }, / payload / null, / signature / h'02d227ed...ccd3774f' ]) ]]> The response contains the Receipt for the Signed Statement. Fresh Receipts may be requested through the resource identified in the Location header.

Status 303 - Registration is running In cases where the registration request is accepted but the Transparency Service is not able to produce a Receipt in a reasonable time, it MAY return a locator for the registration operation, as in this non-normative example: ]]> The location MAY be temporary, and the service may not serve a relevant response at this Location after a reasonable delay. The Transparency Service MAY include a Retry-After header in the HTTP response to help with polling.

Status 400 - Invalid Client Request The following expected errors MUST be returned when the corresponding conditions are encountered. Implementations MAY return other errors, so long as they are valid objects.

Query Registration Status This resource lets a client query a Transparency Service for the registration status of a Signed Statement they have submitted earlier, and for which they have received a 303 or 302 - Registration is running response. Request: Response: One of the following:

Status 302 - Registration is running Registration requests MAY fail, in which case the Location MAY return an error when queried. If the client requests (GET) the location when the registration is still in progress, the TS MAY return a 302 Found, as in this non-normative example: ]]> The location MAY be temporary, and the service may not serve a relevant response at this Location after a reasonable delay. The Transparency Service MAY include a Retry-After header in the HTTP response to help with polling.

Status 200 - Asynchronous registration is successful Along with the receipt the Transparency Service MAY return a locator in the HTTP response Location header, provided the locator is a valid URL. >, / unprotected / { / proofs / 396 : { / inclusion / -1 : [ <<[ / size / 9, / leaf / 8, / inclusion path / h'7558a95f...e02e35d6' ]>> ], }, }, / payload / null, / signature / h'02d227ed...ccd3774f' ]) ]]> The response contains the Receipt for the Signed Statement. Fresh Receipts may be requested through the resource identified in the Location header. As an example, a successful asynchronous follows the following sequence: TS Client <-- 303 Location: .../entries/tmp123 --- TS May happen zero or more times: Client --- GET .../entries/tmp123 --> TS Client <-- 302 Location: .../entries/tmp123 --- TS Finally: Client --- GET .../entries/tmp123 --> TS Client <-- 200 (Receipt) --- TS Location: .../entries/final123 ]]>

Status 400 - Invalid Client Request The following expected errors MUST be returned when the corresponding conditions are encountered. Implementations MAY return other errors, so long as they are valid objects.

Status 404 - Operation Not Found If no record of the specified running operation is found, the Transparency Service returns a 404 response.

Status 429 - Too Many Requests If a client is polling for an in-progress registration too frequently then the Transparency Service MAY, in addition to implementing rate limiting, return a 429 response: { / title / -1: \ "Too Many Requests", / detail / -2: \ "Only requests per are allowed." } ]]>

Resolve Receipt Request: Response:

Status 200 - OK If the Receipt is found: >, / unprotected / { / proofs / 396 : { / inclusion / -1 : [ <<[ / size / 9, / leaf / 8, / inclusion path / h'7558a95f...e02e35d6' ]>> ], }, }, / payload / null, / signature / h'02d227ed...ccd3774f' ]) ]]>

Status 404 - Not Found If there is no Receipt found for the specified EntryID the Transparency Service returns a 404 response: not known \ to this Transparency Service" } ]]>

Privacy Considerations The privacy considerations section of applies to this document.

Security Considerations

General Scope This document describes the interoperable API for client calls to, and implementations of, a Transparency Service as specified in . As such the security considerations in this section are concerned only with security considerations that are relevant at that implementation layer. All questions of security of the related COSE formats, algorithm choices, cryptographic envelopes, verifiable data structures and the like are handled elsewhere and out of scope for this document.

Applicable Environment SCITT is concerned with issues of cross-boundary supply-chain-wide data integrity and as such must assume a very wide range of deployment environments. Thus, no assumptions can be made about the security of the computing environment in which any client implementation of this specification runs.

Authentication Authentication is out of scope for this document. Implementations MAY authenticate clients, for example for the purposes of authorization or preventing denial of service attacks. If Authentication is not implemented, rate limiting or other denial of service mitigations MUST be implemented.

Threat Model

In Scope The most serious threats to implementations on Transparency Services are ones that would cause the failure of their main promises, to wit:

• Threats to strong identification, for example representing the Statements from one issuer as those of another
• Threats to payload integrity, for example changing the contents of a Signed Statement before making it transparent
• Threats to non-equivocation, for example attacks that would enable the presentation or verification of divergent proofs for the same Statement payload

Denial of Service Attacks While denial of service attacks are very hard to defend against completely, and Transparency Services are unlikely to be in the critical path of any safety-liable operation, any attack which could cause the silent failure of Signed Statement registration, for example, should be considered in scope. The impact of DoS attacks can be detected by a client checking that the Transparency Service has registered any submitted Signed Statement and returned a Receipt. Since verification of Receipts does not require the involvement of the Transparency Service, a DoS attack cannot cause the silent loss of a registration. However, this relies on clients actively checking for Receipts and does not prevent the disruption itself. Clients to Transparency Services SHOULD ensure that Receipts are available for their registered Statements, either on a periodic or needs-must basis, depending on the use case. Beyond this, implementers of Transparency Services MUST follow general good practice around defending against network attacks such as flooding, including defenses such as rate limiting.

Eavesdropping Since the purpose of this API is to ultimately put the message payloads on a Transparency Log there is limited risk to eavesdropping. Nonetheless transparency may mean 'within a limited community' rather than 'in full public', so implementers MUST add protections against man-in-the-middle and network eavesdropping, such as TLS.

Message Modification Attacks Modification attacks are mitigated by the use of the Issuer signature on the Signed Statement.

Message Insertion Attacks Insertion attacks are mitigated by the use of the Issuer signature on the Signed Statement, therefore care must be taken in the protection of Issuer keys and credentials to avoid theft and impersonation.

Out of Scope

Replay Attacks Replay attacks are not particularly concerning for SCITT or SCRAPI: Once a statement is made, it is intended to be immutable and non-repudiable, so making it twice should not lead to any particular issues. There could be issues at the payload level (for instance, the statement "it is raining" may be true when first submitted but not when replayed), but being payload-agnostic implementations of SCITT services cannot be required to worry about that. If the semantic content of the payload are time-dependent and susceptible to replay attacks in this way then timestamps MUST be added to the protected header signed by the Issuer. defines COSE header parameters for including timestamp tokens that MAY be used for this purpose.

Message Deletion Attacks Once registered with a Transparency Service, Registered Signed Statements cannot be deleted. Thus, any message deletion attack must occur prior to registration else it is indistinguishable from a man-in-the-middle or denial-of-service attack on this interface.

Use of Unauthenticated HTTP Metadata Implementations that serve multiple application profiles MAY use unauthenticated HTTP-layer signals, such as request headers or distinct registration endpoints, to route incoming Signed Statements to profile-specific processing. However, these signals are not signed, are not committed to the Verifiable Data Structure, and cannot be replayed by Auditors. Implementations MUST NOT use unauthenticated signals as authoritative inputs to the registration decision. Implementations that use such signals for early dispatch MUST ensure that any processing decisions that affect the outcome of registration are fully determined by authenticated inputs, or are otherwise captured in the Verifiable Data Structure, such that the registration process remains deterministic and replayable by Auditors. The authoritative identification of the application profile is carried within the protected header or payload of the Signed Statement, and MUST be verified after signature authentication.

IANA Considerations

Well-Known URI for Key Discovery This section defines the /.well-known/scitt-keys resource in accordance with .

Resource Definition The /.well-known/scitt-keys resource allows clients to discover the public keys used by a Transparency Service to issue Receipts. Clients interact with this resource by issuing an HTTP GET request to /.well-known/scitt-keys. The Transparency Service MUST respond with a COSE Key Set (as defined in ) serialized as application/cbor. The following media types are used with this resource:

• Request: Clients SHOULD include Accept: application/cbor in the request.
• Response: The Transparency Service MUST use Content-Type: application/cbor.

The /.well-known/scitt-keys/{kid_value} sub-resource allows clients to resolve a single public key by its key identifier (kid). Clients interact with this sub-resource by issuing an HTTP GET request. The Transparency Service MUST respond with a single COSE Key (as defined in ) serialized as application/cbor, or a 404 status if no matching key is found. The full normative behavior of these resources, including key lifecycle, error handling, and kid encoding requirements, is defined in and .

Registration Template The following value is requested to be registered in the "Well-Known URIs" registry (using the template from ): URI suffix: scitt-keys Change controller: IETF Specification document(s): RFCthis Status: Permanent Related information:

References Normative References In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings. Fraunhofer SIT Microsoft Research Microsoft Research ARM Traceability in supply chains is a growing security concern. While verifiable data structures have addressed specific issues, such as equivocation over digital certificates, they lack a universal architecture for all supply chains. This document defines such an architecture for single-issuer signed statement transparency. It ensures extensibility, interoperability between different transparency services, and compliance with various auditing procedures and regulatory requirements. This memo defines a path prefix for "well-known locations", "/.well-known/", in selected Uniform Resource Identifier (URI) schemes. In doing so, it obsoletes RFC 5785 and updates the URI schemes defined in RFC 7230 to reserve that space. It also updates RFC 7595 to track URI schemes that support well-known URIs in their registry. Concise Binary Object Representation (CBOR) is a data format designed for small code size and small message size. There is a need to be able to define basic security services for this data format. This document defines the CBOR Object Signing and Encryption (COSE) protocol. This specification describes how to create and process signatures, message authentication codes, and encryption using CBOR for serialization. This specification additionally describes how to represent cryptographic keys using CBOR. This document, along with RFC 9053, obsoletes RFC 8152. The Hypertext Transfer Protocol (HTTP) is a stateless application-level protocol for distributed, collaborative, hypertext information systems. This document describes the overall architecture of HTTP, establishes common terminology, and defines aspects of the protocol that are shared by all versions. In this definition are core protocol elements, extensibility mechanisms, and the "http" and "https" Uniform Resource Identifier (URI) schemes. This document updates RFC 3864 and obsoletes RFCs 2818, 7231, 7232, 7233, 7235, 7538, 7615, 7694, and portions of 7230. This document defines a concise "problem detail" as a way to carry machine-readable details of errors in a Representational State Transfer (REST) response to avoid the need to define new error response formats for REST APIs for constrained environments. The format is inspired by, but intended to be more concise than, the problem details for HTTP APIs defined in RFC 7807. JSON Web Signature (JWS) represents content secured with digital signatures or Message Authentication Codes (MACs) using JSON-based data structures. Cryptographic algorithms and identifiers for use with this specification are described in the separate JSON Web Algorithms (JWA) specification and an IANA registry defined by that specification. Related encryption capabilities are described in the separate JSON Web Encryption (JWE) specification. This document describes the commonly used base 64, base 32, and base 16 encoding schemes. It also discusses the use of line-feeds in encoded data, use of padding in encoded data, use of non-alphabet characters in encoded data, use of different encoding alphabets, and canonical encodings. [STANDARDS-TRACK] This specification defines a method for computing a hash value over a CBOR Object Signing and Encryption (COSE) Key. It specifies which fields within the COSE Key structure are included in the cryptographic hash computation, the process for creating a canonical representation of these fields, and how to hash the resulting byte sequence. The resulting hash value, referred to as a "thumbprint", can be used to identify or select the corresponding key. Informative References This document defines two strategies for handling long lines in width-bounded text content. One strategy, called the "single backslash" strategy, is based on the historical use of a single backslash ('\') character to indicate where line-folding has occurred, with the continuation occurring with the first character that is not a space character (' ') on the next line. The second strategy, called the "double backslash" strategy, extends the first strategy by adding a second backslash character to identify where the continuation begins and is thereby able to handle cases not supported by the first strategy. Both strategies use a self-describing header enabling automated reconstitution of the original content. This document defines two CBOR Object Signing and Encryption (COSE) header parameters for incorporating timestamping based on RFC 3161 into COSE message structures (COSE_Sign and COSE_Sign1). This enables the use of established timestamping infrastructure per RFC 3161 in COSE-based protocols. This document describes the format of a request sent to a Time Stamping Authority (TSA) and of the response that is returned. It also establishes several security-relevant requirements for TSA operation, with regards to processing requests to generate responses. [STANDARDS-TRACK]

Contributors Transmute

United States orie@transmute.industries

Orie contributed examples, text, and URN structure to early version of this draft. Microsoft

United Kingdom amaury.chamayou@microsoft.com

Amaury contributed crucial content to ensure interoperability between implementations, improve example expressiveness and consistency, as well as overall document quality. Business Cyber Guardian

United States dick@businesscyberguardian.com

Dick contributed use cases and helped improve example expressiveness and consistency. MITRE Corporation

United States ramartin@mitre.org

Bob contributed use cases and helped with authoring and improving the document.

stevenlasker@hotmail.com

Steve contributed architectural insights, particularly around asynchronous operations and participated in the initial writing of the document. Microsoft

United States nicolebates@microsoft.com

Nicole contributed reviews and edits that improved the quality of the text.

USA roywill@msn.com

Roy contributed the receipt refresh use case and associated resource definition.