Dense, code-forward body with no padding explaining concepts Claude already knows; terse prose and every code block earns its place, matching the 'lean and efficient; assumes competence' anchor.

Provides fully executable, copy-paste-ready guidance throughout (npm/pip install, complete initialize() bootstrap blocks, the synchronized-wait pattern, a custom Strategy class, expect() assertions, afterAll teardown, and CI yaml).

Clear sequence (Install → Bootstrap → isEnabled → Custom strategy → Rollout determinism → Teardown → Run → CI) with explicit checkpoints: the 'synchronized' wait is flagged as required and reinforced by the Anti-patterns table, plus an explicit afterAll destroy cleanup.

No bundle files are provided (references/scripts/assets absent), so this is a single-file skill organized into clearly signaled sections (When to use, Authoring, Running, CI integration, Anti-patterns, Limitations, References) with no nested reference levels, warranting a top score for a self-contained, well-organized skill.

Enumerates four concrete, distinct testing patterns (bootstrap with a static toggles array, disableMetrics + disablePolling test mode, custom Strategy class + isEnabled assertion, assignment-integrity tests), matching the 'lists multiple specific concrete actions' anchor.

Explicitly answers what ('Wraps Unleash SDK testing patterns: bootstrap... test mode... custom strategy... assignment-integrity tests') and when ('Use when writing tests for code that uses Unleash for feature flags') with an explicit 'Use when' trigger clause.

Uses natural user phrasing ('writing tests for code that uses Unleash for feature flags') plus concrete keywords 'feature flags', 'Unleash', and 'tests', giving good coverage of terms a user would actually say.

Scoped tightly to the Unleash product and feature-flag testing with product-specific triggers, making it clearly distinguishable and unlikely to fire for sibling flag-testing skills.