What problem does it solve? Tagging a release of a .NET library without a structured audit risks shipping silent data corruption, leaked credentials, undocumented public APIs, or stale package metadata. This Skill separates the checks a machine can decide from the reviews that need judgement, so nothing important is skipped and nothing trivial blocks the tag. ## Core Features & Use Cases - Deterministic gates: Release build with warnings-as-errors, unit and integration tests, NuGet package inspection, version marker sweeps, culture-independence tests, wiki sample compilation, link/anchor validation, and secrets scanning. - Judgement reviews: Public API surface diffing, thread-safety guarantees, timeout behavior, cancellation token honoring, nullable annotation audits, NuGet metadata freshness, and trimming/AOT annotation coverage. - Trap catalog: Documents real failure modes from this repository — ignored tests passing filters, anchor slugger bugs, capability flags hiding paths — so green results are interrogated before being trusted. - Use Case: Before tagging tik4net 4.0.0, run the gates, delegate each review to an agent one at a time, and produce a written verdict splitting findings into "blocks the tag" versus "ships with it, tracked". ## Quick Start Ask the assistant to run the pre-release check on tik4net before tagging the next version and report what blocks the tag.
