Anti-Gaming & Sybil Resistance
To protect the integrity of the scoring system, multiple layers of anti-gaming mechanisms are built in from the ground up. Gaming is deterred by making real, sustained responsible behavior the cheapest and most effective way to earn a high score. Attempts to artificially inflate scores are expensive, detectable, and provide only marginal or temporary gains.Key Anti-Gaming Mechanisms
Proof-of-Uniqueness & Sybil Resistance
The Identity Strength component requires progressively stronger signals of being a unique real person.
Basic wallet-only users are capped at lower maximum scores until they add verified channels (email, phone, social, traditional credit, or optional KYC).
Zero-knowledge proofs and decentralized identity attestations prevent one person from controlling multiple high-scoring identities without significant effort and cost.
Future optional integration with proof-of-personhood systems (e.g., Worldcoin-style biometrics) for users who want maximum identity weight.
Behavioral Pattern Analysis
The scoring engine looks for natural, human-like patterns rather than raw volume.
Examples of flagged gaming behavior:
Sudden bursts of tiny circular transfers between owned wallets
Repetitive identical transactions designed to trigger repayment events
High-frequency, low-value activity with no economic purpose
These patterns receive heavily diminished or negative weight in the Behavioral Consistency and Repayment components.
Time Decay & Long-Term Weighting
Recent activity matters, but long-term consistency is required for high scores.
Short-term gaming (e.g., a week of artificial repayments) provides only temporary small boosts that quickly decay.
Sustained, real-world behavior over months and years is required to reach and maintain scores above 80–85.
Economic Disincentives
Many gaming strategies are simply more expensive than the benefit they provide.
Example: Creating dozens of fake identities to farm small boosts costs gas, time, and verification fees far exceeding any lending advantage gained.
Risk Deductions & Anomaly Detection
Unusual volatility, correlated activity across linked wallets, or known sybil cluster patterns trigger deductions in the Risk (R) component.
AML/oracle integrations flag wallets associated with mixers or known gaming farms.
Transparency as Deterrent
Every score change is fully auditable on-chain.
Suspicious patterns are visible to lenders, who can apply their own filters (e.g., “ignore scores with high circular transfer flags”).
This market-level scrutiny further discourages manipulation.
What This Means for Honest Users
Nothing changes, genuine activity (receiving income, repaying loans, normal spending) is rewarded optimally.
You are never punished for legitimate behavior, even if it looks “unusual” in isolation (e.g., large one-time transfers).
Ongoing Improvements: Sybil resistance and anti-gaming are active areas of research. Localcredit can approve new data sources, pattern-detection rules, or weight adjustments as attack vectors evolve, always with transparency and community disclosure. The system is built to make honesty the clearest path to the highest scores and best opportunities.
Last updated