15 skills · sorted by relevance
Audit a Python pytest suite for missed faults, uncovered code, slow tests, and unstable results. Deeper checks deliberately alter code to measure detection, while proposed test removals receive separate validation. The skill reports findings without changing tests.
Review new and existing software tests for the behavior and regressions they actually protect. Identify duplicate or implementation-dependent checks, document evidence before removal, and simplify code used only by tests while retaining meaningful coverage.
Audit an existing test suite to find low-value or duplicate tests and code kept solely for those tests. Receive findings for review, with changes limited to batches you explicitly approve.
Audit a test suite to identify redundant tests, empty checks, and tests tied to internal implementation. Consolidate or remove them with evidence while preserving useful checks that detect regressions, meaning previously fixed problems returning.
Review existing tests and CI, the automated checks run during development, for redundant coverage, misplaced tests, and opportunities to reduce runtime or cost. Receive file-specific recommendations based on documented improvement patterns, with uncertain findings separated.
Evaluate Vitest tests by the observable behavior and realistic regressions they protect. Guide new test authoring and review existing suites for duplication, dependence on implementation details, and code used only by tests, with evidence for retention or removal.
Gate new or changed tests, and audit existing tests for low value, implementation coupling, duplication, and the test-only production seams they keep alive. Use when writing, changing, reviewing, or sweeping tests.
The English summary is being prepared. Showing the original description.
Audit, design, write, materially revise, or remove Tutti tests. Use whenever Codex changes or reviews unit, component, conformance, integration, regression, platform, or repository tests; enforce a protected product contract, credible failure, correct owning boundary, negative-control evidence, overlap review, deterministic setup, and an executing CI lane.
The English summary is being prepared. Showing the original description.
Write or review Windows regression tests for observable contracts, independent expectations, faithful fixtures, and distinct coverage without weakening native proof or architecture guards.
The English summary is being prepared. Showing the original description.
This skill should be used when the user asks to "write a test", "review these tests", "audit tests", "prune low-value tests", or "find duplicate tests", and whenever writing, changing, reviewing, or sweeping tests. Gates new tests and audits low-value, implementation-coupled, or duplicative tests and the test-only production seams they demand.
The English summary is being prepared. Showing the original description.
Gate new or changed Enterprise tests and audit existing tests for observable behavior, credible regressions, duplicate coverage, and test-only production seams.
The English summary is being prepared. Showing the original description.
Decide whether a Pulp test earns its place — before writing it, when reviewing one, or when sweeping a subsystem for low-value, duplicated, or mis-routed tests. Applies a four-question authoring gate, a junk-pattern list, a retention bar, and an evidence template for any deletion, and routes kept tests to the right lane (required gate, pr-fast, slow, nightly). Handles "should I add a test for this", "is this test useful", "audit these tests", "reduce test time", "which tests can go", and "move toward functional tests".
The English summary is being prepared. Showing the original description.
Audit a pytest suite: surviving mutants and untested code, flaky, order-dependent and slow tests, and removal candidates checked on held-out mutants. Report-only.
The English summary is being prepared. Showing the original description.
当 AI 改完代码、要决定这份 diff 放不放行时使用。不逐行精读，按四步走查高风险点——意图是否匹配、有没有碰到边界条件、副作用面有多大、能不能回滚——每步有明确的放过与打回判据。改动文件数、增删行数、动到的迁移/锁文件/CI/配置/鉴权/测试文件和危险 hunk 先由 scripts/diff_risk.py 统计（粘贴进来的 diff 文本也能读，不需要 Git），越过固定阈值时必须人工逐段过。只有口头描述、没有改动内容时不给结论，只列要补的材料。diff 里新写的测试可不可信交给 ai-generated-test-auditor；没有基线的整段新代码走 ai-code-onboarding-checklist。
The English summary is being prepared. Showing the original description.
当 AI 写完测试、要决定信不信这份测试时使用。查三类假信心：断言与实现是否同源复制、有没有边界与异常用例、失败信号够不够强；再规划 2–5 个突变抽样，能执行时再实跑，看把被测代码改坏之后测试会不会变红，没实跑就不声称测试抓得住错。测试全绿不等于代码对，也可能是测试根本抓不住错。要决定整份 diff 能不能合并时走 ai-diff-review-protocol。
The English summary is being prepared. Showing the original description.