Conciseness

Lean and telegraphic throughout — every sentence instructs (e.g., "A regression test that never demonstrably failed proves the mock, not the fix") with zero padding and no explanation of concepts Claude already knows.

Actionability

Fully executable commands with real flags: `pnpm --filter eve exec vitest run --config vitest.<tier>.config.ts <path>`, `pnpm build:compiled`, `pnpm guard:invariants`, `pnpm fmt/lint/typecheck`, `git diff --check`, `git commit -s`, plus concrete checklists (four authoring-gate questions, candidate-evidence fields).

Workflow Clarity

Modes are sequenced up front and the audit flow is ordered Discovery → Retention → Candidate evidence → Edit shape → Validation → Landing → Handoff, with a numbered validation section of exact commands, error-recovery gates ("a missing field means the candidate is not ready for deletion", "Never edit source or tests while Vitest is running"), and checklists for complex processes.

Progressive Disclosure

Good section structure and clearly signaled one-level-deep references ([CAMPAIGN.md](CAMPAIGN.md) for campaign mode, root AGENTS.md, gh-pr-description skill), but everything else lives inline in one dense ~200-line file (the 17-item junk-patterns list is reference-grade material) and CAMPAIGN.md is not present in the provided bundle.

Specificity

Lists multiple concrete actions: "Authoring gate for new tests", "audit workflow for low-value, slow, implementation-coupled, or duplicative tests", and "the test-only production seams they demand" — comprehensive coverage of the skill's capabilities with no vague filler.

Completeness

Explicitly answers both parts: the what ("Authoring gate for new tests plus audit workflow for low-value, slow, implementation-coupled, or duplicative tests...") and the when ("Invoke whenever writing, changing, reviewing, or sweeping tests in the eve repository") with a concrete trigger clause.

Trigger Term Quality

Natural phrases like "writing, changing, reviewing, or sweeping tests" plus "low-value", "slow", and "duplicative" map well to what users would say, but common synonyms such as "flaky", "redundant", "test cleanup", or "prune tests" are missing — good coverage, a few natural terms absent.

Distinctiveness Conflict Risk

The eve-repository scoping and audit quality-bar framing give it a clear niche, but "writing, changing... tests" is a broad trigger that overlaps with general test-authoring skills, so it is mostly distinct with minor overlap risk rather than minimal.