LetspingLETSPING
GET STARTED →

Vercel AI SDK×PostgreSQL

Add Human-in-the-Loop to Vercel AI SDK for PostgreSQL security.

Stop your autonomous agents from making highly destructive, hallucinated API calls to PostgreSQL. Wrap your execution logic in LetsPing to enforce a hard pause until a human reviews the payload.

Start Securing AgentsRead Docs

How it works

1. Intercept PostgreSQL Logic

Instead of hitting the PostgreSQL API directly from your Vercel AI SDK agent, you yield the payload to LetsPing. The agent's execution thread halts completely.

2. Human Review & Patching

You get a push notification on your phone. You review the exact PostgreSQL JSON payload the agent wanted to send. You can approve it, reject it, or patch the values directly.

3. Resume Execution

Once approved, the Vercel AI SDK execution thread resumes, and the actual PostgreSQL mutation is allowed to proceed using the (potentially patched) payload.

Vercel AI SDK Example Implementation
import { LetsPing } from "@letsping/sdk";

const lp = new LetsPing();

const decision = await lp.ask({
  service: "vercel ai sdk-agent",
  action: "postgresql_execute",
  priority: "high",
  payload: {
    id: "123",
    action: "delete"
  }
});

if (decision.status === "APPROVED") {
  // Actually hit the PostgreSQL API now
  await client.execute(decision.payload);
}