Give your agent a phone number.
Your voice agent gets rate-limited. Your browser agent fails KYC. Your autonomous system can't receive a 2FA code. Every agent shipping today is fighting infrastructure that was built for a different kind of user.
A +1+1 fixes that in one call.
One primitive. Five properties. Issued in seconds.
Every +1+1 your agent is issued comes with:
- A real phone number that can place calls, receive SMS, complete verification flows, and operate in every system that already expects a phone number. At the network layer, it is a real phone number.
- A verifiable identity bound to your agent's wallet and to the hardware it runs on. Other systems can confirm the number belongs to an agent (not a spoofed human), that it belongs to youragent specifically, and that the substrate hasn't been tampered with.
- Mission scope enforced before the agent acts You set when, where, on which networks, for what purposes, with what spending limit. The constraints live in the number itself and are checked at every network attach, not in software the agent could be tricked into rewriting.
- A clean revocation path If something goes wrong (compromise, misbehavior, end of task) you burn the number. Connectivity terminates. The audit record is immutable. Captured devices yield nothing reusable.
- An accountability trail Every issuance, modification, and termination is a permanent record. When your enterprise customer or your regulator asks “what did this agent do, on whose authority, under what scope,” you have an answer.
What it'll look like.
The full SDK is in development. The shape is roughly this:
import { Alignent } from '@alignent/sdk'
// Issue a +1+1 for your agent
const agent = await Alignent.issue({
operator: 'acme-corp',
attestation: await getTPMAttestation(),
scope: {
validUntil: '2026-12-31T23:59:59Z',
regions: ['US', 'CA'],
purposes: ['outbound-voice', 'sms-verification'],
spendingLimit: { amount: 100, currency: 'USD' },
networks: ['commercial-5g'],
},
})
console.log(agent.number)
// +1+1-555-0123-4567
console.log(agent.identity)
// did:alignent:0x7a3f...
// Your agent can now place calls, receive SMS,
// and complete verification flows.
agent.onSMS(({ from, body }) => {
const code = extractVerificationCode(body)
// Your agent handles 2FA the way it was always supposed to.
})
// When the task is done, or something goes wrong:
await agent.burn()
// Connectivity terminates in seconds.
// The audit record is permanent.
The interface above is illustrative, not final. Endpoints, parameter names, and the attestation flow will evolve based on feedback from design partners.
Toggle the constraints. Watch which calls survive.
The same scope object that ships with your +1+1, rendered live. The network-attach simulator on the right shows what gets accepted under your current rules.
Regions
Purposes
Networks
Hours
Daily spending limit · $100
await Alignent.issue({
operator: 'acme-corp',
attestation: await getTPMAttestation(),
scope: {
regions: ['US', 'CA'],
purposes: ['outbound-voice', 'sms-verification'],
networks: ['commercial-5g'],
hours: '9–17 PT',
spendingLimit: { amount: 100, currency: 'USD' },
},
})Network-attach simulation
- Outbound call → US carrier, 14:02 PTaccept
- Inbound SMS code, 14:04 PTaccept
- Outbound call → MX carrier, 22:30 PTdeny
- Data session → satellite uplinkdeny
If you're building any of these, we want to talk.
Voice agents
That need to place outbound calls and survive STIR/SHAKEN, anti-robocall filtering, and the rules that are coming next.
Browser and computer-use agents
That need to receive SMS codes, complete verification, and operate in systems that demand a phone number to function.
Autonomous systems and fleets
Drones, robots, delivery vehicles, IoT devices acting on their own behalf, that need identity, scope, and revocation at the network layer rather than retrofitted in software.
Multi-agent systems
Where each agent needs its own identity, its own audit trail, and its own off switch, without the orchestration code becoming the only thing standing between you and a runaway.
Design partners are being onboarded now.
A working prototype of the +1+1 issuance flow won a recent agentic-AI hackathon. It demonstrates the core loop: agent attestation, identity minting, scoped provisioning, and cryptographic burn. The production SDK, sandbox, and developer documentation are in active development.
Design partner spots are limited and weighted toward teams already shipping agents to real users. If that's you, the list is open.
Building something specific and want to talk before the list opens? hello@alignent.com