This document describes a threat model for the
Recognized
Entities specification. It identifies and analyzes security and
privacy threats specific to ecosystems that use
RecognizedEntityCredential resources to express recognition of
entities and the actions they are authorized to perform, such as issuing or
verifying verifiable credentials.
This threat model is a work in progress. It is published alongside the Recognized Entities specification to inform implementers and ecosystem designers of the security and privacy considerations relevant to deployments of the specification. Feedback and proposed additions are welcome via the issue tracker.
Threat modeling is a vital part of specification development. This document provides a threat model for the Recognized Entities specification, which defines a data model for expressing that one or more entities are recognized to perform specific actions — such as issuing or verifying verifiable credentials — within a particular ecosystem.
This threat model follows the methodology described in the W3C Threat Modeling Guide and classifies threats using the STRIDE taxonomy. Threats are grouped into four categories:
Each response to a threat follows one of four strategies, abbreviated ERTA:
Each threat is also tagged with the kind of harm it causes: security, privacy, or market competition. STRIDE has no class for market competition harms, so those threats are classified by their effect on the excluded party.
Not every element in the data flow diagram is listed by a threat. This is by design: the model covers the threats the Working Group chose to analyze, and an element is listed by a threat only when that threat applies to it.
The Recognized Entities ecosystem enables a [=E1|list publisher=] — such as an
individual, a club or organization, a national business register, an
accreditation authority, or an industry consortium — to issue a
RecognizedEntityCredential ([=O1|O1=]) asserting that one or more
[=E2|issuers=] are recognized to perform specific actions. A [=E3|holder=] can
acquire a [=O2|dependent verifiable credential=] from such an issuer and
[=F4|present=] it to a [=E4|verifier=] alongside the
[=O1|recognized entity credential=], allowing the [=E4|verifier=] to confirm
both the integrity of the presented credential and the recognized status of its
issuer.
The core challenge is that a [=E4|verifier=] receiving a [=O2|dependent verifiable credential=] from an unknown [=E2|issuer=] must be able to determine, in an automated and cryptographically-verifiable way, whether that [=E2|issuer=] is recognized within the relevant ecosystem. The Recognized Entities specification addresses this by providing a data model for publishing recognition information as a [=O1|recognized entity credential=], enabling a [=E4|verifier=] to accept [=O2|dependent verifiable credentials=] from [=E2|issuers=] described by [=E1|recognized entity list publishers=].
For this threat model, we consider a minimal instantiation of the Recognized Entities ecosystem consisting of four roles: the [=E1|List Publisher=] (E1) who issues [=O1|recognized entity credentials=], the [=E2|Issuer=] (E2) who receives recognition and issues [=O2|dependent verifiable credentials=], the [=E3|Holder=] (E3) who receives and presents those credentials, and the [=E4|Verifier=] (E4) who validates the credential chain against a configured set of trusted list publishers.
Note: We do NOT model potential attackers as first-class elements, as over-characterizing attackers can lead to analysis bias that is better avoided.
The data flow diagram below illustrates the key elements and data flows in a minimal Recognized Entities ecosystem. The diagram shows how a recognized entity credential (O1) is issued by a list publisher (E1) to an issuer (E2), how the holder (E3) [=F4|presents=] a dependent verifiable credential (O2) alongside the recognized entity credential to a verifier (E4), and how the verifier [=F5|retrieves the published list=] from publisher storage (S1) to confirm the issuer's recognized status. The diagram is deliberately kept at this level of detail; the Dictionary describes the steps within each process.
The following notation is used in the diagram description and dictionary:
| E1 List Publisher |
An entity — such as an individual, club, organization, accreditation authority,
national business register, or industry consortium — that issues recognized
entity credentials that assert that one or more entities are recognized to
perform specific actions. The list publisher is the authority for any recognized
entity credentials it issues.
Defined by: Verifiable Credentials Data Model v2.0 Section 1.2: Ecosystem Overview; Section 3.4: RecognizedEntityCredential. |
| E2 Issuer |
An entity — such as an individual, club, university, trade exporter, or a
conformity assessment body — that has been granted a recognized entity
credential by a list publisher and uses it to issue dependent verifiable
credentials to holders. This entity controls an identifier document, such as a
CID or a DID, whose identifier appears as the credentialSubject.id
of the recognized entity credential and as the issuer of the
dependent verifiable credentials it produces.
Defined by: Verifiable Credentials Data Model v2.0 Section 4.7: Issuer; Section 3.2: RecognizedEntity. |
| E3 Holder |
A person or organization that receives a dependent verifiable credential from an
issuer and presents it — optionally, together with the recognized entity
credential — to a verifier. The holder is not necessarily the credential
subject; an importer, for example, presents an invoice that an exporter issued.
The holder stores credentials in a software application, such as a digital
wallet, and controls what is disclosed to verifiers.
Defined by: Verifiable Credentials Data Model v2.0 Section 1.2: Ecosystem Overview. |
| E4 Verifier |
An entity — such as an individual, organization, customs authority, trade
finance lender, or a market surveillance authority — that receives a credential
presentation from a holder and must determine whether the issuer of that
credential is recognized within the relevant ecosystem. The verifier is
configured with one or more URLs that identify the list publishers, and lists,
that it considers authoritative.
Defined by: Verifiable Credentials Data Model v2.0 Section 1.2: Ecosystem Overview; Section 4: Algorithms. |
| P1 Create List |
The process by which a list publisher vets an issuer to determine whether
they qualify for inclusion in a list, and, if so, creates and
cryptographically signs a new RecognizedEntityCredential,
applying the publisher's private key material to produce a verifiable proof.
The output of this process is a signed [=O1|recognized entity credential=],
delivered to the issuer through [=F2|Issue Recognized Entity Credential=] and
passed to [=P2|Publish List=] for publication in [=S1|Publisher Storage=].
Defined by: Section 3.4: RecognizedEntityCredential; ecosystem governance (applicant vetting). |
| P2 Publish List |
The process by which a list publisher makes a signed
RecognizedEntityCredential publicly available, typically
via HTTPS, so that verifiers can retrieve it. In a private ecosystem, the
publisher restricts who can retrieve it, and confidentiality then depends on
every recipient; see Harms by
Association.
Defined by: Section 3.2: RecognizedEntity; deployment (hosting). |
| P3 Configure Verifier |
The process by which a verifier is configured with one or more recognized entity
list publishers — the URLs or DIDs of list publishers whose recognized entity
credentials the verifier will accept as authoritative, or recognized entity
lists.
Defined by: Section 4.1: Credential-based Discovery; Section 4.2: Identifier-based Discovery; Section 4.3: Recognition Validation; deployment (recognition anchor configuration). |
| P4 Credential Issuance |
The process by which an issuer signs and delivers a dependent
verifiable credential to a holder. This process uses the issuer's private
key material to produce a cryptographic proof over the credential. The process
can optionally also result in the delivery of a recognized entity credential
to the holder.
Defined by: Verifiable Credentials Data Model v2.0 Section 4.2: Verifiable Credentials; Verifiable Credentials Data Model v2.0 Section 4.12: Securing Mechanisms. |
| P5 Credential Presentation |
The process by which a holder assembles and transmits a
credential presentation — containing the dependent verifiable credential
and optionally a recognized entity credential — to a verifier.
Defined by: Verifiable Credentials Data Model v2.0 Section 4.13: Verifiable Presentations; Section D.3: Holder-Based Delivery. |
| P6 Validate Credential(s) |
The process by which a verifier checks the cryptographic proof of a presented
dependent verifiable credential, discovers the recognized entity credentials
that might recognize its issuer, and validates the resulting recognition chain.
Discovery takes one of three forms. Credential-based discovery follows
recognizedIn references and retrieves each list through
[=F5|Retrieve List=]. Identifier-based discovery resolves the issuer's
identifier to an identifier document, selects the WhoisService
declared there, and retrieves the presentation published at its service
endpoint; the issuer controls both the document and the presentation. In
holder-based delivery, the holder supplies the credentials through
[=F4|Present Credential(s)=]. Recognition validation then checks each link of
the chain, the recognition anchor at its end, and the action and schema the issuer
is recognized for.
Defined by: Verifiable Credentials Data Model v2.0 Section 7.1: Verification; Section 4.1: Credential-based Discovery; Section 4.2: Identifier-based Discovery; Section 4.3: Recognition Validation. |
| P7 Update List |
The process by which a list publisher modifies an existing
RecognizedEntityCredential — for example to add, remove, or
revoke recognized entities — and re-publishes the updated credential.
Defined by: Section 3.4: RecognizedEntityCredential; deployment (hosting); ecosystem governance (revocation decisions). |
| F1 Identify Entity |
The data flow by which an issuer provides identity and vetting information
to the list publisher's [=P1|Create List=] process, so that the publisher
can determine whether to include the issuer in a recognized entity
credential. The list publisher vets this evidence under its governance
framework; see Admission on False
Evidence.
Defined by: ecosystem governance (application and evidence). |
| F2 Issue Recognized Entity Credential |
The data flow by which a list publisher delivers a signed
RecognizedEntityCredential to an issuer. This credential
asserts that the issuer's DID is recognized to perform one or more
specific actions within the relevant ecosystem.
Defined by: Section 3.4: RecognizedEntityCredential. |
| F3 Issue Credential(s) |
The data flow by which an issuer delivers a signed dependent verifiable
credential to a holder. The credential contains the issuer's DID as its
issuer value.
Defined by: Verifiable Credentials Data Model v2.0 Section 4.2: Verifiable Credentials. |
| F4 Present Credential(s) |
The data flow by which a holder presents a dependent verifiable credential
— and optionally a recognized entity credential — to a verifier. This
flow crosses a security boundary between the [=C3|holder's system=] and the
[=C4|verifier's system=].
Defined by: Verifiable Credentials Data Model v2.0 Section 4.13: Verifiable Presentations; Section D.3: Holder-Based Delivery. |
| F5 Retrieve List |
The data flow by which [=P6|Validate Credential(s)=] requests, and receives in
response, resources referenced by a recognized entity credential: the list
named by a recognizedIn property, the files named by a
credentialValidator property, and other referenced resources. They
come from [=S1|Publisher Storage=] or other endpoints the list publisher
controls, and every retrieval is subject to the verifier's retrieval policy.
The verifier uses this flow to confirm an issuer's current status or to obtain
a more recent credential than the one provided by the holder. The single arrow
in the diagram represents both the request and the response.
Defined by: Section 4.1: Credential-based Discovery; retrieval policy in Section 4: Algorithms. |
| S1 Publisher Storage |
The persistent storage within the [=C1|publisher system=] where recognized
entity credentials are kept and made available for retrieval. This store is
managed by the list publisher and is typically accessible to the public
via HTTPS. In a private ecosystem, access to this store is restricted.
Defined by: deployment (hosting). |
| S2 Holder Storage |
The persistent storage within the [=C3|holder's system=] where dependent
verifiable credentials and recognized entity credentials are kept. This
storage is accessible only to the holder's software within the
[=C3|holder system=] boundary.
Defined by: deployment (holder software, such as a digital wallet). |
| O1 Recognized Entity Credential |
A cryptographically-signed data object issued by a list publisher to an
issuer. It asserts that the issuer's DID is recognized to perform one or
more actions — such as issuing or verifying specific types of verifiable
credentials — within the ecosystem governed by the list publisher.
Defined by: Section 3.4: RecognizedEntityCredential. |
| O2 Dependent Verifiable Credential |
A cryptographically-signed data object issued by a recognized issuer and
delivered to a holder. The credential asserts claims about its credential
subject, which can be an entity other than the holder, such as a product
described by a certificate of conformity, or a commercial invoice that an
exporter issues and an importer presents. The credential is signed with the
issuer's private key; the holder stores it and presents it to verifiers. Its
trustworthiness depends on the issuer holding a valid recognized entity
credential (O1), making it a "dependent" verifiable credential.
Defined by: Verifiable Credentials Data Model v2.0 Section 4.2: Verifiable Credentials; Verifiable Credentials Data Model v2.0 Section 4.8: Credential Subject. |
| C1 Publisher System |
The system boundary — typically a web server or DID-linked resource host —
managed by the list publisher. It runs the [=P1|Create List=],
[=P2|Publish List=], and [=P7|Update List=] processes and maintains
Publisher Storage (S1) from which recognized entity credentials are served.
Defined by: deployment. |
| C2 Issuer System |
The system boundary — typically a server or service operated by the issuer —
that manages the issuer's credentials, keys, and DID. It runs the
[=P4|Credential Issuance=] process and holds the recognized entity
credential issued to the issuer.
Defined by: deployment. |
| C3 Holder System |
The system boundary — typically a mobile or desktop application — that manages
the holder's credentials, keys, and DID. All credential storage (S2) and
[=P5|Credential Presentation=] logic runs within this system boundary, which is
under the holder's control.
Defined by: deployment. |
| C4 Verifier System |
The system boundary — typically an automated verification service or
human-operated review system — that runs the [=P3|Configure Verifier=] and
[=P6|Validate Credential(s)=] processes. It is configured with recognition
anchors identifying authoritative list publishers.
Defined by: deployment; Section 4: Algorithms. |
A list publisher is an entity that issues
RecognizedEntityCredential resources. List publishers are
the authority for any recognized entity credentials they issue. They
range from national government bodies (such as business registers and
accreditation authorities) to industry consortia and standards organizations.
List publishers have a strong interest in ensuring that only properly vetted entities appear in their lists, and that their list credentials are not forged or misused. They also bear operational responsibility for the availability and integrity of the credentials they publish.
An issuer is any person or organization that has been granted a
RecognizedEntityCredential by a list publisher and uses
that recognition to issue dependent verifiable credentials to holders.
Examples include universities issuing academic credentials, national
business registers issuing trade entity credentials, and conformity
assessment bodies issuing product certificates.
The issuer's primary goal is to issue dependent verifiable credentials that will be accepted by verifiers within the relevant ecosystem. They want confidence that their recognized status is accurately represented, that their recognition credentials are not misappropriated or tampered with, and that their DID remains under their control.
A holder is a person or organization that receives dependent verifiable credentials from issuers and presents them to verifiers. In the cross-border trade use case, the holder is an importer presenting an exporter's invoice credential to customs. In the education use case, the holder is a student presenting a university-issued diploma.
The holder's primary goal is to present credentials that will be accepted by verifiers, without disclosing more information than necessary. Holders have a strong interest in privacy: they do not want their credential presentations to leak information about their activities to list publishers or other third parties.
A verifier is an entity that receives a credential presentation from a holder and must determine whether to accept it. Verifiers range from automated customs clearance systems processing thousands of trade documents per hour to human-operated market surveillance authorities checking product conformity certificates.
The verifier's primary goal is to confirm that the credential it has received was issued by an entity that is genuinely recognized within the relevant ecosystem, and that the credential itself has not been tampered with. Verifiers also have a strong interest in availability: a denial of service against the lists they depend on directly impairs their ability to function.