Переводитъ отвѣты чата на /б/оярскій ѩзꙑ́къ — дореформенная орѳографія (ѣ, і, ъ, ѳ), церковнославянскій складъ рѣчи, аористъ и лурковскій словарь. Use when the user says "боярский", "боярскій языкъ", "говори по-боярски", "переведи на боярский", "/boyar", or asks to render chat/output in mock Old-Russian meme style. Persists for the rest of the session until told to stop.

Create or improve accessible modal dialogs in frontend applications, including semantics, focus management, keyboard behavior, inert background content, form errors, and verification. Prefer native HTML dialog when it fits; use this for implementation and review, not for generic visual styling.

Use when the user asks for Partner Training from supplied source materials. Trigger language: partner training; partner onboarding; channel partner training; reseller training. Produce a grounded interactive digital-human course learners can interrupt with voice questions. Do not invent unsupported facts or claim external certification, competence, or real-world completion. With explicit approval, this Skill may install or update the official PersonWise CLI and uses browser OAuth. Other course-creation requests are handled only when the user explicitly asks for them with this Skill.

Use when the user asks for Developer Quickstart from supplied source materials. Trigger language: developer quickstart; API tutorial; SDK tutorial; first successful API call. Produce a grounded interactive digital-human course learners can interrupt with voice questions. Do not invent unsupported facts or claim external certification, competence, or real-world completion. With explicit approval, this Skill may install or update the official PersonWise CLI and uses browser OAuth. Other course-creation requests are handled only when the user explicitly asks for them with this Skill.

Use for any question or task about giving feedback on someone's tests or measuring a team's test quality over time, such as coaching a junior engineer or a quarterly per-author trend, rather than a merge verdict. Scores tests 1-5 on six design axes with explicit anchors and writes growth-framed feedback.

Use when a pull request needs readable coverage feedback rather than one total. Compares current and baseline coverage (LCOV, Cobertura or JSON) and reports per-file line and branch deltas, new files and hidden drops, with a one-line summary for the PR comment.

Use when existing data (datasets, logs, free text) must be scanned for personal data or redacted: Microsoft Presidio analyzer and anonymizer, built-in recognizers for names, emails, cards, SSNs, IBANs and national IDs, custom pattern recognizers, score thresholds and CI gates.

Use when Azure Functions need testing before deploy: running them locally with Core Tools v4 and Azurite, unit tests for .NET isolated, Node v4 and Python v2 handlers, HTTP, queue and timer triggers, admin-endpoint calls and binding checks in local.settings.json.

Use when a team must create or refresh the human relevance judgments that search tests rely on, before running NDCG, MRR or recall evaluations: query sets, grading scales, rater guidelines, inter-rater agreement, Quepid, TREC-style pooling and refresh cadence.

Use when a scanner flags a CVE that analysis shows is not exploitable in your product, and a machine-readable assertion is needed instead of deleting the finding. Writes and validates OpenVEX statements (not_affected, affected, fixed, under_investigation) with vexctl and attaches them to images.

Use when an app has a Print button or print stylesheet that is untested, or printed and PDF copies break in the wrong places. Tests print CSS with Playwright print emulation and page.pdf(): @page size and margins, page breaks, orphans and widows, and print-only rules.

Use for any question or task about Appium mobile UI tests, even a quick fix, including tests that break after an Appium upgrade and suites that pass on Android but fail on iOS: one suite for iOS and Android (native or React Native) with XCUITest, UiAutomator2 and Espresso drivers, capabilities, clients in JS, Python, Java, Ruby or .NET, simulators, emulators and device farms.

Use when a team wants a coordinated bug bash before a release or after a big change instead of an ad-hoc 'everyone test for an hour'. Prepares the charter, data, environments and sign-up, runs roles, a shared board and live triage, scores findings and writes the same-day wrap-up.

Use when a web frontend needs Core Web Vitals (LCP, INP, CLS) checked on every PR, or its first performance budgets drafted. Sets up Lighthouse CI with Google's thresholds at the 75th percentile, per-route budgets by page type, resource-size budgets, report artifacts and PR comments.

Use for any question or task about Promptfoo or adding quick regression checks for prompts or RAG pipelines in CI, including turning written requirements into assertions: promptfooconfig.yaml providers, prompts and tests, deterministic, semantic, model-graded, latency and cost assertions, promptfoo eval and view, and CI gating.

Use for any question or task about catching LLM quality regressions, including checking a new model version or provider against the old one and prompt tweaks that break other cases. Builds a versioned golden dataset, deterministic runs, exact, semantic, judge and rubric checks, and a pass-rate CI gate with a baseline diff across model versions.