@vouchscore/sdk

@vouchscore/sdk

Thin TypeScript client for the Vouch Trust API.

cd packages/sdk && npm install && npm run build

import { createVouchClient } from "@vouchscore/sdk";

// apiUrl below is the current production deployment. A custom domain
// (e.g. api.vouch.dev) is not registered yet — swap this in once it is.
const vouch = createVouchClient({
  apiUrl: "https://agent-trust-tawny.vercel.app/api/v1",
  apiKey: process.env.VOUCH_API_KEY!,
});

const score = await vouch.getWalletScore("0x...");
await vouch.attestX402Payment({
  wallet: "0x...",
  txHash: "0x...",
  resource: "/api/premium/data",
});

Methods: getAgentScore, getWalletScore, getPayeeScore, batchScore, attestX402Payment, createSpendGuard.

SpendGuard — pre-payment policy for agents

Buyer-side counterpart to the score lookups above: before your agent pays someone, ask the guard. It returns an allow/deny decision plus machine-readable reasons — and nothing else. SpendGuard is strictly non-custodial: it never touches keys, funds, signing, or transaction submission. Execution stays with your wallet stack (Coinbase AgentKit, Privy, ...).

const guard = vouch.createSpendGuard({
  maxPerTxUsd: 10,             // deny any single payment above $10
  dailyBudgetUsd: 50,          // deny once today's allowed total would pass $50
  minPayeeScore: 40,           // deny when the payee scores below 40
  blockOnRecommendation: true, // deny when the payee recommendation is BLOCK
});

const decision = await guard.evaluate({ payee: "0x...", amountUsd: 5 });
if (decision.allow) {
  // hand off to AgentKit / Privy / your own signer
} else {
  console.error(decision.reasons); // e.g. ["payee_score_below_min"]
}

How it works:

• All policy fields are optional — set only the rules you want. minPayeeScore / blockOnRecommendation trigger a GET /v1/payees/{address}/score lookup (skipped when a local rule already denied, so no quota is burned on a dead payment). A purely local policy makes no API calls at all.
• Trust-lookup failures fail closed: the decision is deny with reason payee_trust_unavailable.
• Budget reservation is optimistic: once the local rules pass, the amount is reserved before the trust lookup awaits and returned automatically if the trust rules deny — so concurrent evaluate calls within one process cannot race past the daily budget together. If an allowed transfer then fails or is skipped, call guard.release(amountUsd) to give the reservation back.
• The daily budget counter lives in this process's memory (UTC day): it resets on process restart and is not shared across replicas. Treat it as a runaway-agent brake, not an accounting system — persist your own ledger if you need durable budgets.

See examples/agentkit-spend-guard for a runnable demo and an AgentKit integration sketch.

See OpenAPI and x402 integration.