How to Verify Any Smart Contract or AI Agent Before You Transact

#defi #web3 #ai #security

When someone asks "is this DeFi protocol safe?" or "can I trust this AI agent?", there's now a programmatic answer.

Maiat Protocol is a trust infrastructure layer for the agentic economy. It provides on-chain reputation scoring for any EVM address — smart contracts, wallets, and AI agent tokens alike.

What Maiat Does

Maiat aggregates three data sources to produce a trust score (0–100):

On-chain analytics — contract age, transaction volume, deployer history, verified source code
Community reviews — structured ratings from users who have interacted with the protocol
Outcome reports — verified post-transaction reports (success / failed / dispute)

The result is a verdict:

✅ proceed — score above threshold (default 60)
⚠️ caution — borderline, proceed carefully
🚫 block — high-risk, avoid

API Usage

Maiat exposes a public REST API:

# Free tier — up to 10 req/min
GET https://maiat-protocol.vercel.app/api/v1/score/0xYourAddress

# Trust gate verdict (proceed / caution / block)
GET https://maiat-protocol.vercel.app/api/v1/trust-check?agent=0xYourAddress

# x402 payment-gated endpoint (for AI agents, $0.02 USDC / call)
GET https://maiat-protocol.vercel.app/api/v1/trust-gate?agent=0xYourAddress

The trust-gate endpoint implements the x402 payment protocol — meaning AI agents using Coinbase's AgentKit or awal can autonomously pay for and consume trust data without human intervention.

Why This Matters for AI Agents

In 2026, AI agents are transacting on-chain autonomously. An agent executing a swap or deploying capital needs to verify its counterparty the same way a human would check a contract on Etherscan — but programmatically.

Maiat is designed for this pattern:

// Before any agent-to-agent transaction
const verdict = await fetch(
  `https://maiat-protocol.vercel.app/api/v1/trust-check?agent=${counterpartyAddress}`,
  { headers: { 'X-Maiat-Key': 'mk_...' } }
).then(r => r.json());

if (verdict.verdict === 'block') {
  throw new Error('Counterparty flagged as high-risk by Maiat Protocol');
}

How Trust Scores Are Calculated

Signal	Weight	Notes
Contract verified on Etherscan	High	Unverified source = immediate caution
Deployer wallet age	Medium	Fresh wallets = higher risk
Transaction count (90d)	Medium	Low activity = less data
Community review score	High	Weighted by reviewer reputation
Outcome dispute rate	High	% of transactions that ended in dispute

Use Cases

Pre-swap trust gate — verify a protocol before routing a swap through it
AI agent onboarding — screen new agents before allowing them to join a multi-agent workflow
DeFi due diligence — research any protocol before depositing funds
Smart contract auditing supplement — community signal alongside formal audits

Quick Links

Live app: maiat-protocol.vercel.app
GitHub: github.com/JhiNResH/maiat-protocol
x402 ecosystem: Listed on x402.org/ecosystem under Services/Endpoints
Virtuals ACP: Available as trust_gate, trust_score_query, onchain_report, deep_insight_report offerings

Maiat Protocol is built on Base. Trust scores are computed in real-time using on-chain data + Prisma-backed community reviews. No API key required for the free tier.

DEV Community