Proof of Human Intent
A cryptographic protocol that proves a real human approved critical software actions. Built for the AI era.
AI executes. Humans authorize. Machines verify.
How the Protocol Works
Three cryptographic steps to verifiable human approval
Verify Human
Prove you are a unique human via a Proof of Personhood provider
Bind to Action
Cryptographically bind your proof to a specific commit or deployment
Record Attestation
Create a tamper-evident, independently verifiable attestation
Try the Demo
Three cryptographic steps to verifiable human approval
Approval Request
Select Verification Method
Pluggable architecture supporting 9 Proof of Personhood providers
Use Cases
Real-world applications for verifiable human approval
PR Merge Gating
Require verified human approval before merging AI-generated pull requests. Integrates with GitHub Actions.
Deployment Approval
Ensure a verified human authorized every production deployment. Prevent autonomous AI deployments.
Release Signing
Create tamper-evident proof that a human authorized a software release. Auditable and verifiable.
Compliance & Audit
Immutable audit trail proving human oversight of AI-driven changes. Meets regulatory requirements.
Quick Integration
Add human approval verification to your CI/CD pipeline in minutes
# .github/workflows/human-approval.yml
name: Require Human Approval
on:
pull_request:
types: [labeled]
jobs:
verify:
if: github.event.label.name == 'ready-to-merge'
runs-on: ubuntu-latest
steps:
- uses: pohi-protocol/action@v1
with:
world-id-app: ${{ secrets.WORLD_ID_APP_ID }}
required-level: orbArchitecture
Chain-neutral, zero-dependency core protocol