v0.1.0 - Initial public benchmark baseline

ShopPay Audit Benchmark v0.1.0

Initial public baseline release for evaluating whether AI coding agents can find business-logic defects from written product rules.

Included

• SPEC.md business-rule source of truth.
• Intentionally flawed ShopPay service implementation.
• Baseline tests documenting seeded defects.
• BENCHMARK.md audit task and maintainer answer key.
• docs/SCORING.md 100-point scoring rubric.
• examples/audit-report.md sample report format.
• GitHub Actions test workflow.
• Contribution guide, code of conduct, issue templates, PR template, and roadmap.

How to run

npm test

Expected baseline result: all tests pass against the intentionally flawed benchmark implementation.