Test Data and Fixture Architect

The problem

Brittle test suites, flaky CI pipelines, and accidental production data leaks often stem from poorly architected test data. Developers struggle with shared state collisions, slow setup times, and fixtures that become stale as the schema evolves.

What it does

• Designs deterministic factory and builder patterns to replace brittle, hard-coded seeds.
• Audits existing fixture systems for isolation risks, cross-tenant leakage, and privacy vulnerabilities.
• Creates comprehensive strategies for database cleanup, state reset, and multi-tenant isolation.
• Develops reproducible randomness strategies using seed capture for debugging flaky tests.
• Architects specialized data layers for unit, integration, and end-to-end browser testing.

Why this beats prompting it yourself

General LLMs often suggest simple "faker" scripts that lead to referential integrity issues or slow tests. This skill applies a 27-step architectural audit covering everything from monotonic time control to worker-level isolation, ensuring your test data scales with your infrastructure.

Use cases

• Refactoring a legacy test suite plagued by "mystery guest" dependencies and global state.
• Designing a privacy-safe synthetic data strategy to replace risky production database dumps.
• Building scenario-based builders for complex multi-step workflows like checkout or onboarding.
• Optimizing CI performance by redesigning database truncation and parallel worker data isolation.

Known limitations

Does not execute database migrations or run tests directly. Requires a provided schema or existing fixture code to generate implementation-specific recommendations.