Trust Scoring
How TrustGate Scores Addresses
TrustGate scores two address types: wallet addresses via the Oracle, and contract addresses via Token Shield. All scores are on a 0 to 100 scale. Exact formula weights are intentionally not published to prevent gaming.
Wallet Scoring (Oracle)
Used for any EOA (externally owned account) queried through the Oracle.
What Raises a Wallet Score
Contract Deployments. The single most important signal. Deploying contracts onchain separates builders from users. The significance of what you deployed matters — a protocol used by thousands of wallets carries far more weight than an empty or unknown contract. Long-term builders with multiple meaningful deployments score highest.
Transaction History. Wallets with little or no onchain activity score lower. Sustained, long-term transaction history across multiple protocols is a strong positive signal.
Wallet Age. Derived from the timestamp of the oldest outgoing transaction. Older wallets with consistent activity score higher. Very new wallets are capped regardless of other signals.
Contract Interactions. Genuine participation across diverse protocols signals a real ecosystem participant, not a manufactured identity.
USDC Balance. Demonstrated financial presence on Arc contributes positively.
Activity Spread. Consistent activity spread across many months signals long-term presence, not a burst campaign designed to inflate scores quickly.
Tier Bands
| Tier | Score | Who This Is |
|---|---|---|
| LOW | 0 to 39 | New, inactive, or flagged wallet |
| MEDIUM | 40 to 59 | Active user with limited builder history, or any bot flag present |
| HIGH | 60 to 79 | Active developer or sustained long-term participant |
| HIGH_ELITE | 80 to 100 | Serious long-term builders only |
HIGH_ELITE reflects sustained long-term builder behavior across multiple dimensions. Missing any critical condition hard-caps the score below this tier.
A perfect score reflects exceptional long-term ecosystem participation across every signal category simultaneously.
Payment Routing
| Score Range | Recommendation |
|---|---|
| 0 | BLOCKED |
| Low range | TIME_LOCKED |
| Mid range | INSTANT |
| High range | INSTANT_PRIORITY |
Bot Detection
TrustGate detects velocity anomalies, timer-based automation, self-interaction loops, coordinated activity, and synthetic transaction histories. Any single detected signal hard-caps the wallet at MEDIUM tier regardless of all other signals. There are no exceptions.
Contract Scoring (Token Shield)
Token Shield auto-detects the contract type and applies the correct scoring model.
- ERC-20 address → ERC-20 Token Scoring
- Non-ERC-20 contract → Contract Scoring (free, no payment required)
- Not a contract → error directing to Oracle page
ERC-20 Token Scoring
Holder Quality Weighting. Only wallets that purchased the token count. Airdrop recipients are ignored completely. This removes the fake holder count attack vector entirely — rug pulls cannot boost scores by airdropping to thousands of fresh wallets.
Holder weight is based on the buyer's wallet trust tier. High-trust buyers contribute meaningfully. Low-trust buyers contribute very little. Airdrop recipients contribute nothing.
Deployer Credibility. The deployer wallet is scored through the wallet oracle. Their trust tier is applied as a significance multiplier on the token score. Trust propagates from builder to product.
Score Updates. New tokens update more frequently. Established tokens update less frequently — changes carry more significance. A rising score on a new token is a positive signal. A falling score on an established token is a warning.
Non-ERC-20 Contract Scoring
Scored across six dimensions: verification status, contract age, transaction volume, unique interactors, deployer trust, and Arc ecosystem recognition.
Verification is weighted heavily. Unverified contracts are capped regardless of all other signals — publishing source code is a trust signal. Verified contracts can reach the highest tiers.
Contract age matters. A contract running without incident for many months is more credible than one deployed last week.
Unique interactors matter more than raw transaction count. Many different wallets interacting is fundamentally different from one wallet calling the contract many times. Diversity signals organic adoption.
Deployer trust propagates directly to the contract score. A HIGH_ELITE deployer is a strong positive signal. A LOW deployer is a red flag regardless of what the contract does on the surface.
Arc ecosystem recognition rewards contracts genuinely embedded in the Arc DeFi ecosystem.
Contract Flags
Interaction Velocity. Organic adoption is gradual. A contract receiving an unusually high number of transactions shortly after deployment signals coordinated or automated activity. Hard cap at MEDIUM tier.
Unverified Hard Cap. Not a bot flag. A transparency cap. Unverified contracts can still score HIGH but can never reach HIGH_ELITE regardless of all other signals.
Tier Bands (Contract)
| Tier | Score | Description |
|---|---|---|
| LOW | 0 to 39 | Suspicious, very new, or low-activity |
| MEDIUM | 40 to 59 | Limited history or velocity-flagged |
| HIGH | 60 to 79 | Active with real usage (unverified contracts cap here) |
| HIGH_ELITE | 80 to 100 | Verified, established, high-usage, trusted deployer, Arc-native |
Arcscan vs TrustGate
Arcscan provides raw data: deployer address, creation timestamp, verification status, transaction history, address counters, token type classification.
TrustGate decides quality: whether activity patterns indicate bots, how deployer history translates to trust, whether timing signals manipulation, how interactor diversity signals credibility, whether Arc ecosystem connections add legitimacy, and the final score and tier.
Arcscan shows you data. TrustGate tells you whether to trust it.