AIP-1 Protocol · Now in Public Beta

The HTTPS
for AI Agents

Verify identity. Enforce boundaries. Build trust. The first protocol that makes AI agents accountable.

Quick Start
$ pip install aip-protocol
$ aip create-passport --domain yourco.com
  ✓ Agent passport created
  ✓ Ed25519 keypair generated
  ✓ DID: did:web:yourco.com:agents:v1

Built for the agentic economy

Every capability your multi-agent system needs to be trustworthy.

Identity
Ed25519

Cryptographic Identity
& Trust Scoring

Bayesian reputation from verification history.

Safety
<1ms

Emergency Kill Switch
Protocol

Revoke any agent instantly. Zero propagation delay.

The Cage

Boundary
Enforcement

Action allowlists, monetary limits, geo-fencing.

Verification
3 Tiers

Tiered Verification Pipeline

Tier 0 (<1ms) for low-risk. Tier 1 (~5ms) with attestation. Tier 2 (~100ms) for high-value cross-org actions.

Three steps to verified agents

From zero to production-grade agent security in minutes.

01

Create Passport

Generate a cryptographic identity with Ed25519 keypair, action boundaries, and monetary limits.

02

Sign Intents

Every action your agent takes is packaged into a signed intent envelope with proof of authorization.

03

Verify & Trust

Receiving agents verify the intent through our 8-step pipeline. Failed checks = blocked execution.

Looking for design partners

We're onboarding 3 early partners building multi-agent systems. Free Enterprise access during beta. Your feedback shapes the protocol.