Grades a curated list of test methods and produces a PR-ready table with A-F grades and notes.
npx skills add https://github.com/dotnet/skills --skill grade-tests
Please help me install this Agent Skill. Skill: grade-tests Source: https://github.com/dotnet/skills/tree/main/plugins/dotnet-test/skills/grade-tests Command: npx skills add https://github.com/dotnet/skills --skill grade-tests
Reviewers of pull requests need a quick, per-test verdict on the quality of newly added or modified tests, but full suite-wide audit reports are too long to consume as a PR comment. This Skill grades each specified test method individually and outputs a compact table with a letter grade, score band, and one-line justification.
Grade the test methods in OrderTests.cs and produce a per-test grade table for my pull request comment.
High-intent search queries and answers about installing and using this skill.
Provide the skill with the diff hunk, PR identifier, or an explicit list of fully qualified test method names from the pull request. It grades each test on assertion strength, structure, and anti-patterns, then outputs a Markdown table suitable for posting as a PR comment.
It supports .NET (MSTest, xUnit, NUnit, TUnit), Python (pytest), TypeScript/JavaScript (Jest, Vitest), Java, Go, Ruby, Rust, Swift, Kotlin, PowerShell, and C++. Language-specific extension files define the assertion APIs and idiomatic patterns used during scoring.
No, this skill only grades a curated, explicitly provided list of tests and rejects ambiguous requests like grading the whole workspace. For full suite audits, use the test-quality-auditor agent or the test-anti-patterns skill instead.
Three sub-grades (assertion strength, structure and focus, anti-pattern hygiene) are weighted 0.45, 0.25, and 0.30 into a numeric average mapped to A-F bands. The overall grade is capped at the worst sub-grade, so any F dimension forces an overall F.
It should not: exception and error-path tests using constructs like Assert.ThrowsException or pytest.raises are treated as complete on their own and earn an A for assertion strength when they check the promised error condition. Deductions only apply for observable issues like broad exception types or swallowed exceptions.
doodledoodAnalyze code diffs to identify and quantify test coverage gaps.
dynatrace-ossReview pull request diffs and comments to generate a structured review brief.
AndurilCodeChallenge unit tests to detect false positives from implementation-derived oracles.
rabbicseDetect quality problems in test code with line-level smell analysis.
