-
Decentralized Trust Graph DTG
The model, defined by the ToIP/DIF Decentralized Trust Graph Working Group, in which verifiable trust agents (VTAs) are the nodes, verifiable relationship credentials (VRCs) are the edges, and verifiable trust communities (VTCs) group nodes under shared rules. The stack implements this model; the DTG credentials specification is a working draft, and the stack’s coverage is not a conformance claim.
-
Verifiable Trust Agent VTA
A digital agent representing a node in the Decentralized Trust Graph (DTG). It may represent a person, device, community, or software agent. The participant is identified by a Verifiable Identifier (VID); the VTA is its digital representative, not simply an identifier.
-
Verifiable Trust Community VTC
A community entity identified by a VID and represented digitally by a VTA. Its members participate under shared governance rules. In Working Draft 0.6.0, community membership is represented by reciprocal Verifiable Membership Credentials (VMCs).
-
Verifiable Relationship Credential VRC
A credential attesting to one direction of a peer relationship. Two VRCs, one issued by each peer, form the complete relationship edge under Working Draft 0.6.0. A proof check establishes signature integrity under the applicable mechanism; it does not itself establish the truth of every claim.
-
Verifiable Identifier VID
An identifier for a participant in the trust graph. DTG Credentials Working Draft 0.6.0 uses DIDs and does not mandate a DID method. Identifier control alone does not establish real-world identity. The stack documentation describes did:scid:vh and did:peer:4.
-
Decentralized Identifier DID
A
W3C-standard identifier referring to a subject, such as a person, organization, device, or abstract entity. A DID method defines its creation and resolution rules. Cryptographic control of a DID is distinct from evidence about the subject’s real-world identity.
-
DID Document
A document associated with a DID that can describe verification methods, authentication mechanisms, and service endpoints. It is not itself a profile or proof of real-world identity. A verifier uses the relevant verification relationship and method for the particular check.
-
did:peer:4
A
peer DID method documented by the stack for self-certifying identifiers. Its long form carries encoded DID document information; resolution requirements depend on the form available to the resolver.
-
did:scid:vh
A self-certifying DID method documented by the stack, using hash-chained verifiable history, key pre-rotation, and witness attestations. Verifying an update requires the relevant history and applicable witness policy; a self-certifying identifier alone is not evidence of a person’s identity.
-
VidMethod
A DID-method-agnostic interface with create(), resolve(),
verify(), and a supportsRotation flag. Agent and
RelationshipManager depend on this interface rather than any specific DID method,
so new implementations can be dropped in without
changing either module's public API. Concrete implementations include
ScidVhMethod, DidPeer4Method, and RotatingVidMethod.
-
Witness VTC
A Verifiable Trust Community whose members act as witnesses for DID rotation events. Rather than
a hosted watcher network the P2PTS's native rotation mechanism
has each rotation co-signed by M-of-N members of a decentralized Witness VTC, using
the same issueVrc()/verifyVrc() primitive as any other VRC —
the claim type is 'rotation-witness'. This release-specific record model is distinct
from the draft’s reciprocal VRC edges and VSC witness predicate profile. A fixed, documented "genesis" Witness
VTC is shipped for development and early-adopter use; production deployments are expected to
replace or supplement it.
-
Rotation Log
A hash-chained sequence of DID document versions used in key rotation. Pre-rotation commits to a successor key, and witnesses attest to updates under configured thresholds. Verification depends on the relevant history, signatures, and witness policy.
-
R-Card Relationship Card
A verifiable data structure containing human- and machine-readable information about its publisher. R-cards are typically exchanged alongside relationship credentials and may be encapsulated in a verifiable credential. This specification-level definition does not establish credential wrapping, verified updates, or zero-knowledge presentation in the release.
-
Trust Spanning Protocol TSP
The Layer 2 protocol of the P2P Trust Stack. TSP defines the message format, signing,
encryption, and relationship handshake for authenticated messaging between VTAs. TSP is
transport-agnostic — it can run over libp2p, HTTP, Bluetooth, or any other channel. TSP is
inclusive of DIDComm-like concepts but is not based on DIDComm.
-
Out-of-Band Introduction OOBI
A small JSON object containing a VTA's VID and transport address (libp2p PeerID + multiaddrs).
OOBIs are shared out-of-band (via QR code, link, copy-paste) to bootstrap a relationship.
The OOBI is not a secret — it is the equivalent of a business card for the trust graph.
-
Relationship Formation Invitation XRFI
The invitation message in the TSP relationship handshake (wire tag XRFI in TSP Rev 3). The initiating agent asks a peer to establish a relationship. Establishing a protocol relationship does not establish unique personhood.
-
Relationship Formation Acknowledgement XRFA
The acknowledgment message in the TSP relationship handshake (wire tag XRFA in TSP Rev 3). A completed handshake records protocol participation; applications separately evaluate credentials and policy before authorizing actions.
-
Relationship Formation Decline XRFD
The decline message in the TSP relationship handshake (wire tag XRFD, new in TSP Rev 3). It carries the digest of the message being declined so the peer can correlate the refusal; a declined invitation fails loudly rather than silently.
-
TSP_GEN
A documentation symbol for a signed application message without payload encryption. The receiver checks the signature using the relevant verification material. Changing signed content causes verification against the original signature to fail.
-
TSP_GEN_CONFIDENTIAL
A documentation symbol for confidential application messaging. In v0.2.0 confidential messaging uses HPKE-Base; the surrounding signed envelope binds the sender. Compatibility also depends on the cryptographic suite and protocol version.
-
Hybrid Public Key Encryption HPKE
Hybrid Public Key Encryption (HPKE, RFC 9180) combines key encapsulation, key derivation, and authenticated encryption. Its base and authenticated modes have different sender-authentication properties. In v0.2.0 the TSP confidential-messaging path uses HPKE-Base; sender authenticity comes from the surrounding signed TSP envelope, not from the HPKE mode.
-
Composable Event Streaming Representation CESR
An encoding and framing mechanism for data such as keys, signatures, and messages. CESR represents cryptographic material; it does not itself sign, encrypt, authenticate a participant, or make a claim true.
-
Ed25519
A digital signature algorithm documented in the stack’s cryptographic API. Signature verification checks signed content against a verification key; it does not by itself establish a participant’s real-world identity or the truth of signed claims.
-
X25519
A Diffie–Hellman key agreement algorithm documented in the stack’s cryptographic API. Key agreement is distinct from message signing. The TSP confidential-messaging path uses HPKE-Base over the recipient’s X25519 key, or the ML-KEM-768 + X25519 hybrid KEM.
-
ML-DSA-65
A post-quantum digital signature algorithm (FIPS 204). Shipped in v0.2.0 for signing OOBIs, DID rotations, and TSP envelopes alongside Ed25519. Signature verification checks signed content against a verification key; it does not by itself establish a participant’s real-world identity or the truth of signed claims.
-
ML-KEM-768
A post-quantum key-encapsulation mechanism (FIPS 203). The stack ships an ML-KEM-768 + X25519 hybrid KEM for HPKE confidential messaging, using the draft HPKE-PQ KEM id 0x647a (draft-ietf-hpke-pq-05).
-
Verifiable Data Registry VDR
Infrastructure used by a DID method to support identifier operations and resolution. Requirements vary by method. A method’s independence from a ledger does not eliminate the need to obtain the verification material or history required for a particular check.
-
libp2p
A modular peer-to-peer networking library used as the transport layer in the current P2PTS
implementation. libp2p handles TCP connections, noise encryption, yamux stream multiplexing,
and peer discovery. The P2PTS wraps libp2p behind the TspTransport interface
so that higher-level code never imports libp2p directly.
-
Direct Mode
TSP messaging between endpoint agents without a TSP routing intermediary. The website documents Direct Mode in v0.2.0. Its underlying libp2p connection may use transport relay; that does not make the exchange TSP Routed Mode.
-
Routed Mode
A planned TSP mode using protocol intermediaries. It is distinct from libp2p circuit relay at the transport level. Routed Mode is not presented as an implemented v0.2.0 capability.
-
Trust over IP ToIP
A four-layer architecture for decentralized trust infrastructure, defined by the
Trust over IP Foundation.
The P2PTS is organized around the ToIP model: Layer 1 (Trust Support), Layer 2 (Trust Spanning),
Layer 3 (Trust Tasks), and Layer 4 (Trust Applications).
-
DTG core credential types
Working Draft 0.6.0 defines seven core types: VRC (relationship),
VMC (membership), VDC (delegation), VIC (invitation), VPC (persona), VSC (statement), and VAC
(authority). VEC (endorsement) and VWC (witness) are VSC predicate profiles.
See
Layer 3 for the taxonomy and release distinction.
-
Verifiable Delegation Credential VDC
Establishes that a delegate may act in another entity’s name within
specified bounds. It does not itself confer authority. The application must separately establish
whether the delegator is permitted to perform the requested action.
-
Verifiable Authority Credential VAC
Grants its subject permission to act in its own name within a defined
scope. Authority and representation are separate questions; a VAC does not itself appoint the
subject to act in its issuer’s name.
-
Credential verification and reliance
Proof verification checks the credential’s cryptographic evidence.
Applicable validity and status checks are also needed. Revocation does not erase the original
signature; applications evaluate current status and their own rules before relying on claims
or authorizing actions.