Developers > Docs > Hello world integration
Hello world integration
TrigGuard is execution authorization infrastructure for AI agents. This page shows the smallest Express path that enforces deterministic decisions: PERMIT, DENY, or SILENCE.
1Install packages
npm install trigguard
pip install trigguard
2Authorize and verify
import { createTrigGuard } from "trigguard";
const tg = createTrigGuard({
gatewayUrl: process.env.TRIGGUARD_GATEWAY_URL ?? "https://api.trigguardai.com",
apiKey: process.env.TRIGGUARD_API_KEY,
});
const result = await tg.authorize({
surface: "deploy.release",
actorId: "hello-world",
context: { environment: "staging" },
});
console.log("decision:", result.decision);
const trusted = await tg.verify(result.receipt);
console.log("verified:", trusted);
import os
from trigguard import TrigGuard
tg = TrigGuard(
api_key=os.environ.get("TRIGGUARD_API_KEY"),
gateway_url=os.environ.get("TRIGGUARD_GATEWAY_URL", "https://api.trigguardai.com"),
)
result = tg.authorize(
surface="deploy.release",
actor="hello-world",
context={"environment": "staging"},
)
print("decision:", result["decision"])
trusted = tg.verify(result["receipt"])
print("verified:", trusted)
Full walkthrough: First 10 minutes on the developer onboarding path.
3Decision contract
| Decision | Outcome |
|---|---|
| PERMIT | Execution proceeds. |
| DENY | Execution stops. |
| SILENCE | No authorization; execution stops. |
Keep this mapping explicit in integration code. Do not model decisions with alternate terms.
Navigate by protocol surface