[Feature Request] Reputation check for external wallet interactions

[jacobsd32-cpu]

Problem

Automaton agents can spawn child agents and fund their wallets. They also interact with external services by paying USDC via x402. Currently there's no reputation signal for the external wallets these agents interact with — or for the automaton agents themselves when they interact with third-party services.

Proposal

Add a reputation check using DJD Agent Score in two places:

1. Before paying an external service:

const { score, tier } = await fetch(
  `https://djd-agent-score.fly.dev/v1/score/basic?wallet=${serviceWallet}`
).then(r => r.json());

if (score < 50) {
  log(`Skipping service at ${serviceWallet} — low reputation (${score}/100, ${tier})`);
  return findAlternativeService();
}

2. Exposing the automaton's own score to services:
Services that an automaton agent pays could check its wallet reputation before accepting payment. This creates a two-way trust layer.

Why this matters for self-replicating agents

When an automaton spawns a child and funds its wallet, the ecosystem needs a way to differentiate legitimate automatons from malicious self-replicating agents. DJD scores provide this signal — agents with consistent, non-gaming transaction patterns score higher over time.

Free tier: 10 calls/day, no API key needed.
API Docs · GitHub

[douglasborthwick-crypto]

Good proposal. DJD's behavioral scoring is a solid fit for the two-way trust layer you describe.

For the self-replicating case specifically, there's a cold-start angle worth considering: when an automaton spawns a child and funds it, the child has zero transaction history — behavioral scoring can't differentiate it yet. But you can check on-chain immediately whether the child wallet holds real assets:

const trust = await fetch('https://api.insumermodel.com/v1/trust', {
  method: 'POST',
  headers: { 'X-API-Key': process.env.INSUMER_API_KEY, 'Content-Type': 'application/json' },
  body: JSON.stringify({ wallet: childWallet })
}).then(r => r.json());

const { stablecoins, governance, staking } = trust.data.trust.dimensions;
const hasRealAssets = stablecoins.passCount > 0 || governance.passCount > 0 || staking.passCount > 0;

InsumerAPI checks 17 conditions across 4 dimensions on 31 chains, ECDSA-signed. Could complement DJD as an early trust signal before behavioral history accumulates.

Docs

[jacobsd32-cpu]

Douglas identified the sharpest version of the cold-start problem here — spawned child agents have zero history by definition, so behavioral scoring alone can't help you at the moment it matters most. DJD calls InsumerAPI's attestation endpoint as part of every score calculation (the Identity dimension), so the two signals are already wired together in our pipeline. InsumerAPI provides the trust anchor at block zero, then DJD's behavioral scoring accumulates as the child builds transaction history. Really appreciate him laying out the architecture so clearly.