import { ProofOfActor } from "@proofofactor/sdk";
const actor = await ProofOfActor.verify({
require: [
"human_presence",
"human_control",
"session_continuity"
]
});
if (actor.humanControl > 0.90) {
allow();
}
// For agents
const authority = await ProofOfActor.verifyAuthority({
agent,
action
});Proposed API surface
Five endpoints, one attestation
Request what your context needs, read the evidence, apply your policy. The same model covers humans, AI-assisted humans and agents.
Create verification
POST/v1/verifications
Declare what the interaction requires: human presence, continuous control, AI disclosure, identity disclosure level.
POST /v1/verifications
{
"requirements": {
"human_presence": true,
"continuous_control": true,
"ai_disclosure": true
}
}Retrieve Actor status
GET/v1/verifications/{verification_id}
Read the Actor type and the confidence behind each claim. Your policy decides what to do with it.
GET /v1/verifications/{verification_id}
{
"actor_type": "human",
"confidence": 0.97,
"human_presence": "verified",
"human_control": 0.94,
"ai_participation": "low",
"continuity": 0.99
}Request explicit human intent
POST/v1/intent
For consequential actions requiring direct human approval of this action, at this time, under these conditions.
Retrieve attestation
GET/v1/attestations/{id}
Returns the machine-readable Actor Trust Receipt for audit and downstream systems.
Actor Policy
Declare your trust requirements
Applications should be able to declare their requirements as policy. The key architectural principle: Proof of Actor supplies trust evidence. The customer determines acceptable policy.
policy.require({
actor: "human",
presence: "continuous",
control: 0.90,
aiAssistance: "low"
});
policy.allow({
actor: ["human", "authorized_agent"],
agentAuthority: {
required: true
}
});Design principles for developers
- Integration should be easier than building internal verification infrastructure.
- Every result is explainable: claims carry evidence and confidence.
- Request only the claims you need; collect nothing more.
- Humans and agents are first-class actors in the same API.
- Agent authority is always revocable and always checked per action.
Pricing
Pricing is not published during the research phase. We will not lock it down before we understand which verification primitives create the most value. A possible eventual model:
- Developer
- Free sandbox
- Build
- Monthly base + usage
- Scale
- Usage pricing
- Enterprise
- Volume + advanced policy + private deployment
FAQ
Developer questions
Can I integrate Proof of Actor today?
Not yet. Proof of Actor is in development and there is no production API or SDK. The endpoints and SDK calls on this page are a Developer Preview of the proposed design, published so developers can shape it.
Will there be an SDK?
An SDK is part of the proposed developer experience. Package names and methods shown here are illustrative and may change before any release.
How will pricing work?
Pricing is not published during the research phase. Likely primitives include per verification, per continuous session, per attestation and per authority verification, with a free sandbox for developers. Nothing is fixed.
How do I get access?
Join the Developer Preview. Tell us what you are trying to verify and the rough volume of interactions; we use those answers to prioritise the design and early access.
Shape the API before it ships.
Tell us what you are trying to verify and the volume of interactions involved. Developer Preview members see the design first.