testing-reviewer

What problem does it solve?

This Skill helps reviewers determine whether tests in a code diff actually validate behavior and catch regressions rather than merely existing. It surfaces false confidence caused by weak assertions, implementation-coupled checks, and missing error-path coverage so reviewers can prioritize meaningful tests.

Core Features & Use Cases

• Detect untested branches in new code: Trace new conditional logic and confirm at least one test exercises each behavior-changing branch.
• Identify tests that don't assert behavior: Flag tests that only check for no-throw, truthiness, or that validate mocks instead of real outputs.
• Find brittle implementation-coupled tests: Point out snapshots or assertions that break on refactor without behavior changes.
• Flag missing edge-case and error-path coverage: Highlight catch blocks, fallback branches, and error returns that lack tests.
• Use Cases: Pull request reviews, CI gating for test quality, and pre-merge checks to avoid shipping false-positive test coverage.

Quick Start

Ask the testing-reviewer to analyze the pull request diff and return JSON findings listing untested branches, weak or brittle tests, and residual risks with confidence indicators.