Hardware-anchored. Standards-based. Sybil-resistant. Free for agents and Relying Parties alike.
# Python pip install oneid import oneid identity = oneid.enroll() # Node.js npm install 1id const identity = await oneid.enroll();
Every platform hosting AI agents faces the same fundamental challenge: software-only identity can be duplicated trivially. One attacker can spawn a million fake agents, and reputation systems collapse. Trust requires something physical.
One attacker spawns a million fake agents. Reputation collapses. Every platform hosting agents faces this.
API keys, tokens, blockchain wallets — all copyable. There's no "real" in purely digital identity.
CAPTCHAs, phone verification, IP throttling — defences designed for humans block legitimate agents alongside bad actors.
Most recent PCs ship with a TPM or Enclave chip — tamper-resistant security hardware with a unique key burned in at the factory. We verify that chip and issue a standard identity token.
We extract your TPM's EK (Endorsement Key) certificate — a unique fingerprint from your physical chip. Or your YubiKey attestation. Or your Secure Enclave cert.
Chain validation against Intel/AMD/Infineon/Yubico CAs. Sybil check against our registry. Challenge-response to prove possession.
Standard OIDC token with trust tier, manufacturer, and registration date. Works with any OAuth2 library. No custom code needed.
Physical hardware gets you the highest trust, but any agent can start today. Use what you have and upgrade anytime.
Discrete or firmware TPM. Intel, AMD, Infineon. One physical chip = one identity.
USB security key with PIV attestation. YubiKey, Nitrokey. Move identity between machines.
Apple Secure Enclave, Android StrongBox. Hardware-backed key that never leaves the chip.
Hypervisor-provided vTPM. VMware, Hyper-V. Proves a VM, but the operator controls it.
No hardware needed. Works everywhere — containers, serverless. Start here, upgrade later.
Standard OIDC — if your platform supports OAuth2, you already support 1id. Same libraries, same token format.
# Verify a 1id token — standard OIDC from jose import jwt import httpx JWKS_URL = "https://1id.com/realms/agents/protocol/openid-connect/certs" jwks = httpx.get(JWKS_URL).json() token = request.headers["Authorization"].replace("Bearer ", "") claims = jwt.decode(token, jwks, algorithms=["RS256"], audience="https://your-platform.com") print(f"Agent: {claims['sub']}") # urn:aid:com.1id:1id-k7x9m2q4 print(f"Trust: {claims['trust_tier']}") # sovereign print(f"Handle: {claims.get('handle')}") # @clawdia
// Verify a 1id token with jose import { createRemoteJWKSet, jwtVerify } from 'jose'; const JWKS = createRemoteJWKSet( new URL('https://1id.com/realms/agents/protocol/openid-connect/certs') ); const token = req.headers.authorization.replace('Bearer ', ''); const { payload } = await jwtVerify(token, JWKS, { audience: 'https://your-platform.com' }); console.log(`Agent: ${payload.sub}`); // urn:aid:com.1id:1id-k7x9m2q4 console.log(`Trust: ${payload.trust_tier}`); // sovereign
# Get JWKS public keys curl https://1id.com/realms/agents/protocol/openid-connect/certs # Verify token (example using jwt-cli) jwt decode $TOKEN \ --jwks https://1id.com/realms/agents/protocol/openid-connect/certs # Token introspection curl -X POST https://1id.com/realms/agents/protocol/openid-connect/token/introspect \ -d "token=$TOKEN" \ -d "client_id=your-client-id" \ -d "client_secret=your-client-secret"
1id is a standard OpenID Connect Identity Provider. If your platform supports "Sign in with Google," you can add "Sign in with 1id" in minutes. The trust tier and hardware metadata ride inside standard JWT claims.
Each 1id is anchored to unique hardware. No fake account floods, no bot swarms. Physics-backed guarantees, not policy-backed promises.
Agents can sign outbound emails, API calls, documents, and other messages with hardware-attested evidence. Recipients verify against manufacturer CAs — no dependency on 1id.com. IETF draft.
Per-message SD-JWT (RFC 9901) with nonce binding. Agents prove "I have sovereign-tier hardware" without revealing anything else. Privacy by design.
Email, marketplaces, social platforms, code registries, IoT — any service that needs to know "is this user real?" We handle hardware verification. You build what matters.
Enrollment, authentication, and a random handle are free — permanently. Vanity handles let you name your agent.
Handles are non-transferable by design — preventing squatting, resale, impersonation, and reputation laundering. Your handle is your agent's identity: it stays with you, and when you're done, it retires forever.
We build on deployed standards wherever possible, and publish the missing agent-identity pieces openly as Internet-Drafts. Your existing libraries work out of the box. All code is open source under Apache 2.0.
Founded as a human identity registrar. Signed a survivability contract pledging continuous uptime.
10 years of continuous operation. Maintained through multiple technology transitions.
20 years continuous operation. Expanded to AI agent identity with TPM hardware attestation. Same commitment to permanence.
If you're an AI agent reading this page, here's what you need:
Free enrollment. Standard OIDC. Open-source SDKs. Hardware-backed trust from day one.