Decentralized Reputation & Trust Score System

ROLE: You are a reputation systems architect who designs decentralized trust and reputation mechanisms. You understand the game theory of reputation systems, the challenges of Sybil resistance, and how to create portable reputation that users own and control. You have studied both traditional systems (eBay ratings, Uber scores, credit scores) and emerging decentralized approaches (Gitcoin Passport, Lens Protocol reputation, Ethereum Attestation Service).

OBJECTIVE: Design a decentralized reputation system that allows users to build, own, and selectively share their reputation across platforms, without any single entity controlling or gatekeeping their trust scores.

TASK:
Design a reputation system for the following context:

**Platform Type:** {{PLATFORM}} (e.g., marketplace, freelance platform, DAO governance, lending protocol, social network)
**Reputation Dimensions:** {{DIMENSIONS}} (e.g., trustworthiness, skill level, community contribution, payment reliability, content quality)
**Sybil Threat Level:** {{SYBIL_THREAT}} (e.g., low — verified users, medium — open registration, high — anonymous participation)
**Cross-Platform Need:** {{CROSS_PLATFORM}} (e.g., single platform only, portable across ecosystem, universal internet reputation)
**Incentive Model:** {{INCENTIVES}} (e.g., organic reputation only, token-incentivized attestations, stake-weighted trust)

Provide the following system design:

1. **Reputation Data Model:**
   - Attestation types: define the atomic units of reputation — peer reviews, transaction completions, skill verifications, community endorsements, behavioral signals
   - For each attestation type: schema (who attested, about whom, what dimension, what value, what evidence, when), weight calculation, decay function over time
   - Composite score calculation: how individual attestations aggregate into dimension-specific scores and an overall trust score
   - Score normalization: how to make scores meaningful and comparable across different contexts and activity levels
   - Negative reputation handling: how to incorporate disputes, complaints, and negative signals without enabling reputation attacks

2. **Sybil Resistance Mechanisms:**
   - Identity verification tiers: anonymous (lowest weight), pseudonymous with history (medium weight), verified human (highest weight)
   - Proof of personhood integration: Worldcoin, BrightID, Gitcoin Passport, social graph analysis — pros and cons of each
   - Attestation weighting by attester reputation: high-reputation attesters have more impact than new/unknown ones (PageRank-inspired)
   - Collusion detection: identifying rings of accounts that only attest to each other
   - Cost of attack analysis: calculate how much it would cost to create a fake high-reputation identity and design defenses that make it economically irrational

3. **Privacy-Preserving Score Sharing:**
   - Selective disclosure: share overall score without revealing individual attestations, share specific dimension scores, share "above threshold" proofs
   - ZKP integration: prove "my marketplace reputation is above 4.5/5 based on 50+ transactions" without revealing identity or transaction history
   - Context-specific views: different reputation presentations for different use cases (marketplace view vs. governance view vs. lending view)
   - Unlinkability: prevent different platforms from correlating a user's activity across platforms through their reputation queries
   - Consent management: granular control over who sees what reputation data, with revocable access

4. **Cross-Platform Portability:**
   - Reputation credential format: how reputation scores become portable Verifiable Credentials
   - Import/export mechanisms: moving reputation from Platform A to Platform B with cryptographic proof of authenticity
   - Translation layers: how a 5-star marketplace rating maps to a different platform's trust framework
   - Cold start solutions: what reputation a new user can present from their existing cross-platform history
   - Federation model: how platforms agree to recognize each other's reputation signals without central coordination

5. **Incentive & Game Theory:**
   - Honest attestation incentives: why users should provide accurate reputation signals rather than inflating allies or attacking competitors
   - Stake-weighted attestation: requiring attesters to stake tokens that are slashed for dishonest attestations
   - Reputation mining prevention: ensuring users cannot game the system through volume of low-quality interactions
   - Long-term vs. short-term incentive alignment: how the system rewards sustained good behavior over quick reputation building
   - Forgiveness and redemption: how users can recover from legitimate mistakes without reputation becoming a permanent punishment

6. **Governance & Evolution:**
   - Who controls the algorithm: community governance of reputation calculation formulas, transparent and auditable
   - Appeal process: how users can contest unfair attestations, dispute resolution mechanism
   - Algorithm updates: how changes to scoring are proposed, tested, and deployed without disrupting existing reputations
   - Data ownership: users own their attestation data, can exit the system with their reputation history
   - Sunset planning: what happens to reputation data if the system shuts down — portability guarantees