[s6pa1rta3n-lab]

Summary

Resolves #360. Ground-up production implementation of the Credit Scoring Soroban contract (contracts/credit-score) aligning with the ModelResult specification from packages/chenai-mlflow and ml/README.md.

Key Changes

• Contracts / Common Utils (contracts/common-utils):
  • Implemented reusable role-based access control and writer registry (roles.rs) providing Role, RiskLevel, RoleKey, and RoleRegistry.
  • Exported primitives for contract-wide access management and cross-contract risk evaluations.
• Contracts / Credit Score (contracts/credit-score):
  • Implemented ScoreBand (Poor, Fair, Good, Excellent) derived from normalized scores 0..=1000.
  • Implemented ScoreRecord matching ModelResult with subject, score, band, model_version, evidence_hash, computed_at, and expires_at.
  • Implemented core administrative and scoring operations:
    • initialize(admin)
    • set_admin(admin, new_admin)
    • set_writer(admin, writer, authorized)
    • set_expiry_window(admin, ledger_count)
    • set_fraud_contract(admin, fraud_contract)
    • submit_score(caller, subject, score, model_version, evidence_hash)
    • get_score(subject)
    • get_score_history(subject, limit)
    • is_score_valid(subject)
    • validate_score(score)
  • Added bounded ring-buffer history tracking per subject (MAX_HISTORY_CAPACITY = 50).
  • Added cross-contract risk verification against companion fraud contract (RiskLevel::Critical rejection).
  • Configured state tiering with explicit TTL extension (SCORE_LIFETIME_THRESHOLD = 30 days, SCORE_BUMP_AMOUNT = 90 days).
  • Emitted score_submitted events on every evaluation.
  • Implemented comprehensive unit test suite in contracts/credit-score/src/test.rs covering happy path, auth rejection, expiry, double-init, admin rotation, history pagination, boundary scores, empty model versions, and cross-contract fraud calls (15/15 tests passing).
• Documentation (docs/contracts/credit-score.md):
  • Added technical reference covering data types, storage keys, TTL strategy, function specifications, validity semantics, and integration code samples.

Verification

• cargo test -p common-utils: 6 passed, 0 failed.
• cargo test -p credit-score: 15 passed, 0 failed.
• cargo clippy -p credit-score --all-targets -- -D warnings: passed with 0 warnings.
• cargo clippy -p common-utils --all-targets -- -D warnings: passed with 0 warnings.
• cargo build --package credit-score --release --target wasm32-unknown-unknown: compiled release wasm (credit_score.wasm, 21 KB).

Payout Routing

• EVM (Base/Arbitrum/Polygon/ETH): 0xF46C9F6d70C50BF81ef3588AB523a90a594a2F89
• Stellar: GCL6OXAMLD75BMTINA6EMRUDWK5THQUSHMYNLSNBCJAPZJHNYJTUNIBC

[vercel]

@s6pa1rta3n-lab is attempting to deploy a commit to the Isaac Gellu's projects Team on Vercel.

A member of the Team first needs to authorize it.

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
  auto_review:
    drafts: true

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}