farley-scorelisted

Score test-suite quality with Dave Farley's 8 Properties of Good Tests (Understandable, Maintainable, Repeatable, Atomic, Necessary, Granular, Fast, First/TDD) and produce a weighted 0-10 Farley Index with per-property evidence, tautology-theatre and mock anti-pattern detection, the five worst tests, and prioritised recommendations. Read-only: it never edits code. Supports Java/JUnit, Python/pytest/unittest, JS/TS Jest/Vitest, Go testing, and C# NUnit/xUnit. Use whenever someone asks how good their tests are, not just how many: 'score my tests', 'farley score', 'farley index', 'review test quality', 'are these tests any good', 'find useless tests', 'tests that pass no matter what', 'too many mocks', 'test smells', or before trusting a suite's coverage number. NOT for measuring coverage or running tests. NOT for writing or fixing tests - it reports; the fixing is separate work. For learning to spot the problems yourself, use farley-score-coach.
wagneripjr/skills · ★ 0 · Testing & QA · score 70

Install: claude install-skill wagneripjr/skills

# Farley Score: test design review Measure test *quality*, not coverage. Read the suite, detect signals per test method, score each of the eight properties twice (once from static signals, once from semantic judgment), blend the two, and let the bundled calculator do all the arithmetic. The result is a Farley Index from 0 to 10 whose every number points to evidence in the code. Start by reading [host-runtime.md](references/host-runtime.md). It covers path resolution and the Claude Code/Codex differences. **Read-only.** Analyse only. Do not create, edit, rename or delete files in the project. The one exception is writing the report to `farley-score-report.md` at the project root, and only if the user asks for a file. If the user asks for fixes, point at the recommendations and say to apply them test-first. The value of the review depends on the reviewer not also being the author of the change. ## Entry point **Invoked with a target** (a path, or "the whole project"): go straight to the workflow. **Invoked bare**: ask with a menu, not a numbered list in prose. Question: "Welcome to Farley Score! What would you like to do?", header "Farley Score". Options: 1. **What is this?** Explain that the Farley Index measures test quality against the 8 properties and runs 0-10. Understandable and Maintainable weigh 1.5×, because tests that read like specifications and survive refactoring deliver the most long-term value. Fast weighs 0.75×, because a slow, well-designed test beats a