Register an agent. Make a governance call. Get a proof receipt. Five minutes.
npm install @vorionsys/sdk
import { Vorion } from '@vorionsys/sdk';
const vorion = new Vorion({ localMode: true });
// 1. Register an agent
const agent = await vorion.registerAgent({
agentId: 'my-agent-001',
name: 'My First Agent',
capabilities: ['read:*', 'write:documents'],
});
console.log(`Registered: ${agent.getName()} (${agent.getId()})`);
// 2. Check trust score
const trust = await agent.getTrustInfo();
console.log(`Trust: ${trust.score}/1000 — Tier T${trust.tierNumber}`);
// 3. Request a governed action
const result = await agent.requestAction({
type: 'read',
resource: 'documents/report.pdf',
});
console.log(`Allowed: ${result.allowed}`);
console.log(`Proof ID: ${result.proofId}`);
// 4. Report outcome to build trust
if (result.allowed) {
await agent.reportSuccess('read');
}
# With Docker
docker compose -f apps/cognigate-api/docker-compose.yml up --build
# Or run locally
cd apps/cognigate-api && npm run dev
The API starts at http://localhost:3000. Verify with:
curl http://localhost:3000/health
import { Vorion } from '@vorionsys/sdk';
const vorion = new Vorion({
apiEndpoint: 'http://localhost:3000',
});
// Same API as local mode
const agent = await vorion.registerAgent({
agentId: 'my-agent-001',
name: 'My First Agent',
capabilities: ['read:*', 'write:documents'],
});
# Register an agent
curl -X POST http://localhost:3000/api/v1/agents \
-H 'Content-Type: application/json' \
-d '{"agentId":"my-agent-001","name":"My First Agent","capabilities":["read:*"]}'
# Check trust
curl http://localhost:3000/api/v1/trust/my-agent-001
# Submit a governance intent
curl -X POST http://localhost:3000/api/v1/intents \
-H 'Content-Type: application/json' \
-d '{"agentId":"my-agent-001","action":"read","resource":"documents/report.pdf"}'
# Get proof receipt
curl http://localhost:3000/api/v1/proofs/latest?agentId=my-agent-001
git clone https://github.com/vorionsys/vorion.git
cd vorion
npm install
npx tsx examples/quickstart.ts
| Concept | Description |
|---|---|
| Agent | Any AI system registered with Vorion |
| Trust Score | 0-1000 score reflecting an agent's behavioral reliability |
| Trust Tier | T0 (Sandbox) through T7 (Sovereign) — determines permissions |
| Intent | A request from an agent to perform an action |
| Proof | Cryptographic receipt of a governance decision |
| CAR | Canonical Agent Record — unique identity for each agent |