cenconq25

cenconq25/claude-code-app-studio

Skill Claude Code

Quality review of test files and manual evidence docs. Goes beyond existence checks — evaluates assertion coverage, edge cases, naming conventions, and evidence completeness. Verdict: ADEQUATE / INCOMPLETE / MISSING per story. Run before QA sign-off or on demand.

cenconq25/claude-code-app-studio

Skill Claude Code

Detect non-deterministic tests by reading CI run history. Aggregates pass rates per test, identifies intermittent failures, recommends quarantine or fix, and maintains a flaky-test registry. Best run during Polish phase or after a CI run streak.

cenconq25/claude-code-app-studio

Skill Claude Code

Generates a framework-specific test helper library — factories, mocks, custom matchers, and shared assertion utilities — under tests/helpers/. Reduces boilerplate in new test files. Run after /test-setup once the project has a few real tests.

cenconq25/claude-code-app-studio

Skill Claude Code

Scaffold the test framework and CI pipeline for the project. Creates the tests/ directory structure, framework-specific test runner, and GitHub Actions or Bitrise workflow. Run once during Technical Setup before the first sprint.

cenconq25/claude-code-app-studio

Skill Claude Code

Generate a structured user-test (beta) report template OR analyze existing user-test notes into a structured format. Standardizes how qualitative feedback turns into actionable findings tied to PRDs and stories.

cenconq25/claude-code-app-studio

Skill Claude Code

Section-by-section UX spec authoring for a single screen, flow, or HUD element. Reads concept, design bible, parent PRD, and platform conventions to produce design/ux/ .md. Run after the relevant PRD is approved and before stories cite UI behavior.

cenconq25/claude-code-app-studio

Skill Claude Code

Validates a UX spec for completeness, accessibility (WCAG 2.2 AA), platform conventions (iOS HIG / Material 3), and implementation readiness. Produces APPROVED / NEEDS REVISION / MAJOR REVISION verdict with specific blockers. Run in a fresh session — never the same as the authoring session.