What problem does it solve? After editing .NET, C#, or Blazor code, it is easy to claim a change works based on weak evidence — a single focused test, a stale package cache, or a build that silently skipped locked outputs. This Skill enforces evidence-backed verification so every claim of "done" is backed by exact commands and honest pass/fail results. ## Core Features & Use Cases - Four-tier verification ladder: focused RED/GREEN TDD checks, focused module regression sets, full project gates (build, test, format, git diff --check), and ad-hoc verifier scripts for handoff evidence. - Build graph and artifact freshness: reconstructs the build graph from solutions, CI, and package configuration to catch stale NuGet packages, unbundled JS/CSS assets, and consumers resolving old package versions. - Windows lock recovery: diagnoses MSB3021/MSB3027 output-tree locks from IDEs or stale testhost.exe processes and verifies via isolated --artifacts-path directories instead of killing the user's IDE. - Use Case: You changed a Razor component and its tests on Windows, but dotnet test fails with a DLL lock. The Skill identifies the lock, runs build and focused tests in an isolated temp artifacts directory, and reports focused versus full-suite evidence separately. ## Quick Start Ask the agent to verify your recent .NET changes by running the focused tests, full build and test gate, dotnet format check, and git diff check, reporting the exact commands and results.
