Conciseness

Efficient and assumes Claude's competence — no basic-concept teaching — but the skip-when-impractical guidance is stated three times ('Do not force a test when it would be impractical...', 'Do not add tests when the practical signal is weak; use manual or scripted verification', 'Prefer no new test over a bad test') and could be consolidated; not a 5 due to that repetition.

Actionability

Concrete directives throughout ('Add the smallest focused test that would have caught the bug', 'Confirm it fails for the intended reason') plus a concrete fallback list ('targeted script, manual reproduction command, browser automation, snapshot comparison, log assertion, or focused integration check'); not a 5 because 'Run relevant adjacent tests, type checks, lint, or scenario checks' remains generic and what to do when the test still fails after the fix is left implicit.

Workflow Clarity

Clear 7-step sequence with explicit validation checkpoints ('Run the new test before fixing. Confirm it fails for the intended reason') and an explicit feedback loop ('If it passes or fails for an unrelated reason, correct the test or reproduction before editing the implementation'), plus rerun and adjacent-validation steps and an evidence-based final report. Not below 5: checkpoints, error recovery, and reporting are all present.

Progressive Disclosure

The body is under 50 lines, needs no external references, and is organized into well-signaled sections (Workflow, If a Failing Test Is Impractical, Guardrails, Final Response) — each earning its place. This meets the simple-skill exception for scoring 5 with well-organized sections and no bundle files.

Specificity

The description names the domain ('TDD, a failing test, or a regression test', 'an obvious cheap local test target') but never states any concrete action the skill performs — no capability verb appears anywhere. Closest to 'Names the domain but actions are minimal or generic'; not a 3 because no concrete skill actions are stated.

Completeness

'When' is fully and explicitly covered ('Use only when... Skip when...') but the description never states what the skill does — this matches 'only when is present without what'; not a 3 because there is no clear 'what' even weakly implied, and not a 4/5 because the what-half of the question is simply absent.

Trigger Term Quality

Natural user phrases are present and well-chosen ('explicitly asks for TDD, a failing test, or a regression test', 'bug', 'test path'), giving good keyword coverage; not a 5 because common natural synonyms such as 'unit test', 'test-first', or 'write a test' are missing.

Distinctiveness Conflict Risk

The tight gating ('Use only when... OR when the bug has an obvious cheap local test target. Skip when the test path is unclear, expensive, integration-heavy, or not requested') defines a clear niche with distinct triggers and minimal conflict risk — it would not be mistaken for a general testing or debugging skill.