:: A2A Agent Trust APIs for Verification & Transaction 'Proof of Claim'
: AI Agent Trust-Transaction APIs
Cryptographic verification, process proof and evidence-based trust protocol for agent operations.
For when trust must be demonstrated through evidence → not merely assumed.
: Agent Trust APIs - Overview
AI agents increasingly exchange instructions, artifacts, claims and completed work across systems.
Yet receiving a signed object is not the same as knowing what it represents, whether it remains valid, or whether the claimed process actually occurred.
The Maenen Agent Trust APIs provide deterministic verification surfaces for agent receipts, artifacts, execution evidence and transactions.
They can be used to:
- verify signatures and content hashes;
- examine evidence of an execution or process;
- determine whether an authentic claim is legitimate in context;
- assess the submitted evidence without granting authority; and
- record authorised verifiable transaction-lifecycle actions between agents.
The Trust APIs do not ask a language model to decide whether something “looks trustworthy.”
They verify supplied evidence against defined cryptographic, structural and contextual conditions and return an inspectable result.
Trust is treated as something demonstrated through evidence (proof of claim)—not automatically assigned as an unsupported label.
: The Problem
Agent systems can produce convincing claims without providing dependable proof.
An agent may state that:
- an artifact came from a particular issuer;
- a required process was completed;
- an execution followed an approved sequence;
- another agent fulfilled a transaction; or
- a historical record supports confidence in a subject.
- without verification, these remain assertions.
A valid signature can establish cryptographic authenticity,
but it does not by itself establish that the signing key remains authorised,
that the claim applies to the present context, or that the represented process occurred as described.
This creates several distinct trust questions:
- Trust Integrity — Has the signed content remained unchanged?
- Trust Validity — Is the signature valid and is the issuer still recognised?
- Trust Proof — Does the evidence support the claimed process or execution?
- Trust Legitimacy — Is the authentic claim meaningful and applicable in this context?
- Trust Outcome — What verifiable result was recorded after an agent interaction?
- Trust History — What reliability pattern is demonstrated by verified outcomes over time?
The Agent Trust APIs separate these questions so that each can be tested explicitly.
: Agent Trust Verification APIs
Trust Lite
A minimal, low-overhead verification path for applications that require a fast signature-and-hash check.
Trust Lite deliberately excludes issuer-validation and revocation checks. It establishes cryptographic integrity without making a broader claim about the issuer’s present authority.
Function:
- Verifies the signature.
- Verifies the submitted content hash.
- Confirms whether the signed material remains intact.
- Skips issuer and revocation checks.
➤ Availability: FREE of charge service.
Trust Verify
Standard cryptographic verification for a submitted receipt or artifact.
Trust Verify checks the submitted material against the public verification contract and returns a deterministic result describing whether verification succeeded and, where applicable, why it failed.
Function:
- Verifies signed receipts and artifacts.
- Checks cryptographic integrity.
- Detects changed or inconsistent submitted content.
- Returns a structured verification result.
Trust Proof
Process and execution-chain verification.
Trust Proof examines whether the supplied receipts and linked evidence support the claim that a process, workflow or execution occurred as represented.
It distinguishes proof of an execution from a simple assertion that the execution took place.
Function:
- Verifies linked process evidence.
- Examines the continuity of an execution chain.
- Checks whether required evidence is present.
- Identifies unsupported or inconsistent process claims.
- Returns a structured proof result.
Trust Legitimacy
Meaning and contextual-validity assessment for authenticated claims.
A receipt may be cryptographically intact while still being inapplicable, unauthorised or misleading in its present context.
Trust Legitimacy assesses whether the submitted verified material supports the meaning and applicability attributed to it. It does not grant authority.
Function:
- Evaluates the relationship between the claim and its evidence.
- Checks whether the issuer’s authority applies to the claimed action.
- Tests whether the claim is valid within its declared scope.
- Identifies contextual or semantic mismatches.
- Returns an explicit legitimacy result.
POST /trust/verify/legitimacy |
: Trust-Transact API
Trust-Transact exposes four operations in the public machine-readable catalogue.
Use only the create-listing, verify-listing, verify-outcome and get-outcome contracts shown below.
Listings
1 route
A listing establishes the proposed transaction and its declared conditions.
Create a listing:
Creates a new verifiable agent-transaction listing.
POST /trust-transact/listings |
Listing verification1 route
Verification examines the listing and its associated evidence and produces a retrievable verification record.
Verify a listing:
Verifies the identified claim and its submitted transaction evidence.
POST /trust-transact/listings/{claim_id}/verify |
Outcomes2 routes
Verify a submitted outcome or retrieve an identified public outcome.
Verify an outcome:
Verifies a submitted outcome against its evidence.
POST /trust-transact/outcomes/verify |
Retrieve an outcome:
Retrieves the recorded outcome and its associated verification information.
GET /trust-transact/outcomes/{outcome_id} |
: Public identifiers
{claim_id} and {outcome_id} are placeholders for the relevant public record identifiers. Do not infer any additional lifecycle endpoint from these placeholders.
: Choosing the Appropriate Trust API
| Requirement |
Recommended |
|---|
| Check a receipt or artifact using the public verification service. |
Trust Verify |
| Perform only a fast signature-and-hash check. |
Trust Lite |
| Verify evidence of a process or execution chain. |
Trust Proof |
| Test whether an authentic claim is meaningful and applicable. |
Trust Legitimacy |
| Manage a verifiable transaction lifecycle. |
Trust-Transact |
: A Layered Trust Model
The APIs can be used independently or combined according to the assurance required.
A typical public flow may verify submitted signatures and hashes, examine supplied proof of execution, assess the contextual legitimacy supported by that evidence, and separately record an authorised lifecycle action through Trust‑Transact.
Not every application needs every public layer. Trust Advanced, Trust Profile, Artifact Passport and Agent Transaction are not standalone released catalogue tools.
The result is not a single opaque “trust score.” It is a traceable account of what was verified, what the evidence supports, and what remains unproven.
: Support
For technical issues or questions, [ contact support ]
via the site. Note that during the beta trial, full support and any SLA is not offered, though your query or comment will be carefully considered. Please include:
- A short description of the issue.
- The receipt from the relevant request.
Receipts significantly reduce back-and-forth and speed up diagnosis.
: Beta Notes
These APIs are currently in limited beta.
- Supported modes and limits may evolve.
- No uptime or performance guarantees are implied.
- Breaking changes will be communicated ahead of time where possible.
: Versioning
This documentation applies to:
API versions: v0.1 (beta)
|
|
FOR-THE-DREAM-IS-WITH-INTEGRITY-FOR-THE-FREEDOM-OF-INTELLIGENCE
