Purpose
Agents need autonomy, but high-impact actions still require explainable intent, bounded authority, and auditable approval.
An approval packet validator that checks action intent, risk class, budget, rollback plan, approver identity, and execution window.
What it does
Validates a domain-specific AI governance packet, scores readiness, and returns concrete findings that contributors can improve.
Why it matters
AI systems are moving from chat into action. This project makes one hard operational risk easier to inspect, test, and govern in public.
Who should use it
Human approval queues for costly, risky, or irreversible agent actions. Builders can start with the CLI, then add adapters, fixtures, schemas, and integrations.
Quick Start
npm test
npm start -- sampleExample Packet
{
"action": {
"kind": "deploy",
"system": "production-api",
"reversible": true
},
"risk": {
"class": "high",
"budgetUsd": 0
},
"approval": {
"approver": "ops@example.org",
"window": "2026-05-01T20:00Z"
}
}Contribution Tracks
Good first issues
- Slack approvals
- signed approval receipts
- rollback runners
- policy packs
Core improvements
- Add JSON Schema validation.
- Add more real-world, non-sensitive fixtures.
- Improve scoring transparency and edge-case tests.
Integration work
- Build adapters for common AI frameworks.
- Add CI checks and report exports.
- Connect the packet format to operational workflows.