grade-tests

Grades a specified set of test methods individually and produces a concise table mapping each test (fully-qualified name) to a letter grade (A–F), a score band, and a one-line note — designed to be posted as a PR comment. Use when the caller wants per-test feedback on a curated list of methods (for example, the new or modified tests in a pull request), not a suite-wide audit. Polyglot: .NET, Python, TS/JS, Java, Go, Ruby, Rust, Swift, Kotlin, PowerShell, C++. Input is a list of test methods (or method bodies / file+line spans); output is a compact markdown table plus a short summary. DO NOT USE FOR: full suite audits (use test-quality-auditor agent or test-anti-patterns), writing new tests (use code-testing-generator agent or writing-mstest-tests), fixing failures, or measuring code coverage.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A thorough, highly actionable grading skill with a clear sequenced workflow and explicit validation; its main weaknesses are redundancy between the scoring rules and the trailing Validation/Pitfalls sections, and a large inline rubric with no progressive disclosure into its own reference files.

Suggestions

Collapse the Common Pitfalls table into the Step 3 scoring rules it restates, or convert it into a short 'quick-reference' that points back to the sub-grade sections, to remove the conciseness redundancy.

Move the Critical/High/Medium/Low anti-pattern catalog and the sub-grade combination rules into a references/ file (e.g. grading-rubric.md) and keep SKILL.md as an overview that links one level deep, improving progressive disclosure.

Trim or merge the Validation checklist so each item is either a genuinely new check or a pointer to its source rule, rather than a restatement.

Dimension	Reasoning	Score
Conciseness	The body is domain-specific grading logic (not concepts Claude already knows), but the Validation checklist and Common Pitfalls table substantially restate rules already specified in Steps 3–4 (e.g., exception-only tests, always-true literals, Go/Rust table-driven loops), which could be tightened.	2 / 3
Actionability	Fully concrete guidance — exact sub-grade pattern tables, the formula "0.45 × Assertion + 0.30 × Anti-pattern + 0.25 × Structure", letter/band thresholds, the ≤120-char note limit, the 50-row cap, and a copy-paste-ready markdown table format.	3 / 3
Workflow Clarity	A clear Step 0 (reject ambiguous input) through Step 5 (report) sequence with an explicit validation checklist for this batch operation, satisfying the feedback-loop requirement for batch grading.	3 / 3
Progressive Disclosure	Well-sectioned with clear headings (not a monolithic wall) and it defers language-specific content to the extensions skill, but the large core grading rubric lives entirely inline with no reference/scripts/assets bundle of its own, so content that could be split is kept inline.	2 / 3
	Total	10 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A high-quality description: third-person voice, concrete actions, an explicit Use-when trigger, and strong distinctiveness via a DO NOT USE FOR routing clause. Every dimension hits the top anchor with no verbosity penalty.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "Grades a specified set of test methods individually and produces a concise table mapping each test (fully-qualified name) to a letter grade (A–F), a score band, and a one-line note" — rather than vague language.	3 / 3
Completeness	Explicitly answers both what (grade tests → table with grade/band/note) and when ("Use when the caller wants per-test feedback on a curated list of methods"), with an explicit Use-when clause as required for a 3.	3 / 3
Trigger Term Quality	Covers natural user phrasing — "per-test feedback on a curated list of methods", "the new or modified tests in a pull request", "PR comment" — that a reviewer would actually say.	3 / 3
Distinctiveness Conflict Risk	A clear niche (per-test grading of a curated PR list) with explicit disambiguation — "DO NOT USE FOR: full suite audits (use test-quality-auditor agent or test-anti-patterns), writing new tests…" — making conflict with sibling skills unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: dotnet/skills
Path: plugins/dotnet-test/skills/grade-tests/SKILL.md
Commit: 9b27066

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.