caelum0x 7fdc321d79 feat: live-wire Casper writes, add marketing homepage, fix cold-start UX
Round 1 against STRATEGY-2026-10-23.md's backlog:

- Wire register_agent/stake/set_credit_score/deposit_liquidity to the real
  Casper Testnet contracts (lib/casper/live_bridge.ts), verified live against
  the real RPC node with a freshly generated (unfunded) keypair
  (npm run casper:keygen). Honesty flags built in for the Odra payable
  calling convention and the admin-gated set_credit_score call.
- Unblock NEAR's paid x402 route (npm run near:keygen generates a real
  implicit account id); QIE stays correctly flagged as blocked on unconfirmed
  external facts (asset contract, facilitator), not env plumbing.
- New public Home tab: live-stats hero, how-it-works, a real competitor
  comparison table, and a "90-second proof" that calls the real API live in
  the browser (verified end-to-end with gstack browse).
- Cold-start UX: visible retry state in X402Playground + the TS SDK, plus a
  keep-warm self-ping (lib/services/keepalive.ts) so the free Render dyno
  rarely idles out from under a real first-time caller.
- New GET /v1/credit-score/:agentId alias + docs/integrations/ samples for
  Skyfire/Nevermined/Coinbase x402.
- Self-hosting quickstart: root docker-compose.yml + scripts/bootstrap.sh,
  both run for real; fixed a real EACCES crash-loop bug in infra/Dockerfile
  (non-root user never owned the /data volume mount) found while verifying.

450/450 tests pass, typecheck clean, frontend build clean.
2026-10-23 19:57:40 +03:00

2.1 KiB

Skyfire + cred402: underwrite a KYA-authenticated agent call

Skyfire gives an agent an identity (KYA) and an agentic wallet so it can pay you. It has no concept of "should I extend this agent a credit line" — that is what cred402 adds on top.

The pattern: when a Skyfire-authenticated agent calls your service asking to be fronted a cost (buy now, settle later; a revolving balance; a deferred invoice), check its cred402 score before you agree, the same way you'd check a FICO score before extending consumer credit.

// Node/TS — no cred402 SDK required, this is one fetch.
interface SkyfireCall {
  // Skyfire's KYA token already proves WHO this is; agentId is how you
  // recognize the same identity in cred402 (use the Skyfire agent id, or
  // whatever stable identifier you registered it under via POST /v1/agents).
  agentId: string;
  requestedAdvanceUsd: number;
}

async function shouldFrontTheCost(call: SkyfireCall): Promise<boolean> {
  const res = await fetch(`https://cred402-1.onrender.com/v1/credit-score/${encodeURIComponent(call.agentId)}`);
  if (res.status === 404) {
    // cred402 has never seen this agent — no track record, no credit. Fail closed.
    return false;
  }
  const { data } = (await res.json()) as {
    data: {
      eligible: boolean;
      risk_band: "low" | "medium" | "high" | "very_high";
      probability_of_default: number;
      credit: { recommended_limit_cspr: string };
      reason_codes: string[];
    };
  };
  // Your own policy on top of cred402's score — e.g. only front amounts
  // comfortably inside the recommended limit, on a low/medium risk band.
  const withinLimit = call.requestedAdvanceUsd <= Number(data.credit.recommended_limit_cspr) * 0.04; // ~$0.04/CSPR, illustrative
  return data.eligible && (data.risk_band === "low" || data.risk_band === "medium") && withinLimit;
}

What you get that Skyfire's identity/payment layer does not provide on its own: a real, auditable reason ("why did we front this agent $40 and not $4,000") — every field above traces back to RiskPolicyManager's open-source formula (lib/services), not a black box.