Best Practices Compliance
Best practices compliance covers the engineering fundamentals that keep codebases healthy and maintainable — testing strategy, code quality, error handling, and architectural separation. These checks apply broadly to any recipe that produces or modifies code.
Applicability
This category applies to any recipe or guideline that defines, generates, or modifies source code. If a recipe produces implementation artifacts, these checks govern how that code is tested, structured, and maintained.
Checks
unit-test-coverage
Business logic MUST have unit tests with meaningful assertions.
Applies when: recipe implements business logic or utility functions.
Guidelines:
test-pyramid
Test distribution SHOULD follow the test pyramid (many unit, fewer integration, fewest E2E).
Applies when: recipe defines a test strategy or test suite.
Guidelines:
atomic-commits
Changes MUST be committed atomically — one logical change per commit.
Applies when: recipe describes a development workflow or commit strategy.
Guidelines:
code-linting
Code MUST pass configured linting rules without suppression.
Applies when: recipe produces source code in any language.
Guidelines:
post-generation-verification
AI-generated code MUST pass build, test, lint, and log verification before acceptance.
Applies when: recipe involves AI-assisted code generation.
Guidelines:
explicit-error-handling
Errors MUST be handled explicitly; MUST NOT be silently swallowed.
Applies when: recipe includes error paths, exception handling, or Result/Optional unwrapping.
Guidelines:
- Fail Fast (principle)
separation-of-concerns
Business logic MUST be separated from presentation and infrastructure.
Applies when: recipe spans multiple architectural layers (UI, domain, data).
Guidelines:
- Separation of Concerns (principle)
good-test-properties
Tests MUST be fast, isolated, repeatable, and self-validating.
Applies when: recipe defines or includes test implementations.
Guidelines: