Turns a thin or ambiguous story into a reviewable test list - a backlog item that is a short paragraph plus the click-through support recorded for themselves, a spec that is mostly a list of accepted formats, or a tech design pasted into the ticket while the last few releases still shipped missed cases. Takes the story or feature spec and emits a markdown test-case matrix, one row per case (id, title, precondition, steps, expected, tier), covering happy path, alternate paths, boundaries, and negative paths, before any test code is written. Output is the human-reviewable matrix that goes into TestRail / Qase / Xray, not Gherkin scenarios. Use when a story needs its cases enumerated and agreed before automation starts.
日本語の概要は準備中です。原文の説明を表示しています。
実行スコア 未検証
インストール方法を見るインストールする前に、エージェントに与えられる指示の中身を確認できます。
A user story is not a test plan. The "three amigos" hand-off from product to QA needs an intermediate artifact: a human-reviewable test-case matrix that enumerates the cases the story implies, before anyone writes Gherkin or test code. This skill produces that matrix.
The matrix is what manual testers paste into TestRail / Qase / Xray. It is also the contract that the downstream automation engineer reads when picking up the story - gherkin-from-stories (in qa-bdd) needs a stable AC layer, and the AC layer is derived from this matrix, not the other way around.
The PractiTest 2026 State of Testing Report finds 70% of teams already use AI for test-case creation but only 40.7% report achieving "more diverse and complex test cases" (https://www.practitest.com/state-of-testing/) - the dominant failure mode is shallow output (one happy path × three trivial restatements). This skill exists to constrain the output shape so that the matrix is forced through equivalence-partitioning, boundary, and negative-path lenses.
Do not use this skill when:
gherkin-from-stories or manual-test-script-author.Read the user story (and any attached AC, mockups, or rejection notes) and extract:
If any of (1) - (5) is missing from the story, stop and request clarification. A matrix derived from incomplete input will mask the gap rather than surface it. ISTQB's test analysis step is explicit that test conditions must trace to identified test bases; an unclear story is not a test basis.
For each (action, object) pair, generate cases by walking three lenses (equivalence partitioning, boundary value analysis, decision table testing):
For each input parameter, identify the valid and invalid equivalence classes. One representative case per class.
| If parameter is… | Valid classes | Invalid classes |
|---|---|---|
| String with format constraint (email, URL, phone) | Conformant | Empty / wrong format / null / oversized |
| Numeric with range | Min ≤ x ≤ max | x < min, x > max, NaN, negative, zero (if not in range) |
| Enum | Each documented value | Undocumented value, mixed case |
| Reference (foreign key, file path) | Existing | Missing, deleted, belongs-to-other-tenant |
For numeric / length-bounded parameters, add cases at min, min-1, max, max+1. Skip if the parameter is unbounded (free-text comment, etc.).
If the story implies decision logic (role-based access, plan tier, feature flag, A/B variant), add one case per condition combination. Use a decision table to be exhaustive; collapse rows that have the same outcome.
Output is a markdown table. Required columns:
| Field | Notes |
|---|---|
| ID | <story-key>-TC-<n>, e.g. CART-142-TC-03. Stable across iterations. |
| Title | Imperative single sentence: "Adds a product to the cart as an anonymous user". No "should". |
| Tier | smoke / regression / edge / negative. Tier rationale belongs in the next column. |
| Precondition | One sentence; references fixtures by name where possible. |
| Steps | Numbered. Declarative (per Cucumber better-gherkin guidance) - describe behavior, not UI mechanics. |
| Expected | One sentence per observable post-condition. |
| Source claim | Which sentence in the story / AC this case traces to. Forces traceability. |
Story: "As an anonymous shopper, I want to add a product to my cart, so that I can check out without signing in first."
| ID | Title | Tier | Precondition | Steps | Expected | Source claim |
|---|---|---|---|---|---|---|
| CART-142-TC-01 | Adds an in-stock product to an empty cart | smoke | Anonymous session; product SKU-001 in stock | 1. Open product page for SKU-001. 2. Add to cart with default qty. | Cart count = 1; product line shows SKU-001. | Story sentence 1 |
| CART-142-TC-02 | Adds a second product to a cart that already has one | regression | Anonymous session; cart already contains SKU-001 | 1. Open product page for SKU-002. 2. Add to cart. | Cart count = 2; both SKUs present. | Implied by "add" semantics |
| CART-142-TC-03 | Rejects adding an out-of-stock product | negative | Anonymous session; SKU-099 is out of stock | 1. Open product page for SKU-099. 2. Attempt to add to cart. | Add-to-cart control disabled or 409 returned. Cart count unchanged. | Story constraint (implicit) |
| CART-142-TC-04 | Rejects adding when cart is at the per-session limit | boundary | Anonymous session; cart has the maximum number of items per the documented MAX_CART_ITEMS | 1. Open any product page. 2. Attempt to add to cart. | 409 / cart-full message. Cart count unchanged. | Constraint: MAX_CART_ITEMS |
| CART-142-TC-05 | Persists cart across page refresh | regression | Anonymous session; cart has 1 item | 1. Refresh the page. | Cart still shows the same item. | Implied by "without signing in" |
The matrix is the authoring artifact. The next steps in the workflow are:
gherkin-from-stories (in qa-bdd) which converts the rows the team wants to automate into Gherkin scenarios.The matrix is plain markdown. Common integrations:
ID → TestRail ID, Tier → Section, the rest to standard fields.cases/bulk endpoint or the Qase CLI; the markdown table maps row-by-row.docs/test-cases/<story-key>.md for review alongside the implementation PR.| Anti-pattern | Why it fails | Fix |
|---|---|---|
| One row per UI step ("user clicks button", "user types email") | Matrix becomes a script, not a case list. ISTQB calls these procedures, not cases. | One row per case; UI mechanics live in manual-test-script-author (in qa-manual-testing). |
| Five rows that all assert the same happy-path outcome with cosmetic variation | Same equivalence class repeated; doesn't add coverage. | One row per equivalence class (Step 2 Lens 1). |
| No negative cases at all | This is the PractiTest 2026 "test factory" failure mode. | Mandate at least one negative-tier row per story unless the story has no error contract. |
| Source-claim column blank or "Story" | Blocks traceability; if the story changes, you can't tell which rows need re-review. | Cite the specific sentence / AC bullet. |
Cases that depend on implementation choices ("user clicks the React <AddToCartButton> component") | Implementation-coupled cases churn with refactors. | Declarative phrasing per Cucumber better-gherkin. |
| Matrix produced from a story missing actor / postconditions | The matrix masks the gap; downstream sees noise. | Step 1 explicitly halts on missing fields. |
model-based-test-graph-author (in qa-ai-assisted).gherkin-from-stories (BDD path), manual-test-script-author (manual execution path), or ai-test-generator (code generation path).gherkin-from-stories.manual-test-script-author.ai-test-generator. Always pair with a curation and shallow-coverage-review pass downstream.negative-test-generator.boundary-value-generator.まだレビューはありません。使ってみた感想をお寄せください。
ログインなしで書き込めます。使ってみた感想、質問、動かなかった環境などを気軽にどうぞ。
まだコメントはありません。最初のコメントを書いてみませんか。
Builds a CI gate that fails the build on **new** WCAG / a11y violations introduced by a PR while grandfathering pre-existing violations on a per-rule / per-page baseline. Aggregates verdicts from axe-core / pa11y / Lighthouse a11y / WAVE / IBM Equal Access scans. Use when a project has accumulated a11y debt and a strict "zero violations" gate would block every PR - the ratchet pattern lets the team ship while preventing regressions.
日本語の概要は準備中です。原文の説明を表示しています。
Workflow skill that builds an A/B-test validity checklist from an experiment proposal, walking the canonical design-correctness gates - pre-registered OEC/power/guardrails, randomization unit + SRM check, assignment integrity, telemetry, peeking discipline, novelty/primacy, post-experiment SRM re-check - into a per-experiment checklist + sign-off form. Use when launching, auditing, or governing an experiment. For pitfall mechanics (guardrails, peeking) see experiment-results-interpreter's references; to read an already-valid result use experiment-results-interpreter; for per-SDK harness tests use experiment-sdk-testing - this gates DESIGN, not SDK code.
日本語の概要は準備中です。原文の説明を表示しています。
The single install-flow skill: the reference contract for the PWA install flow (installability gate fields, `beforeinstallprompt` handshake, per-platform paths - Android WebAPK / iOS Share menu / Firefox no-op - and the `display-mode` post-install signal, in references/install-flow-reference.md) plus the build-an-X workflow that emits the Add-to-Home-Screen suite. Walks the four-stage timeline, emitting one test per gate cell per [web.dev/articles/install-criteria][install-criteria], the deferred-prompt → `prompt()` → `userChoice` chain per [web.dev/articles/customize-install][customize-install], the iOS Safari manual-metadata branch (`apple-touch-icon`) per [web.dev/learn/pwa/installation][learn-pwa], and the post-install `(display-mode: standalone)` MQ assertion. Output: a Playwright spec with per-stage cells plus a coverage matrix. Use when a PWA's manifest, icons, or install handler change, when install conversion drops at an unknown stage, or when triaging a flaky install assertion.
日本語の概要は準備中です。原文の説明を表示しています。
Build-an-X workflow that uses an LLM to map existing tests to spec sections - given a spec doc + the test suite, the LLM identifies which tests cover which sections, surfaces uncovered sections (gap), and recommends specific tests to add. Output is a coverage matrix per spec ID. Scope is mapping tests that already exist and naming the gaps, not authoring tests for new acceptance criteria. Use when a spec doc and a test suite both exist but nobody can say which requirements are actually covered - before a release sign-off, an audit, or a decision about where to spend the next round of test effort.
日本語の概要は準備中です。原文の説明を表示しています。
Generates tests from natural-language specs (acceptance criteria, user stories, requirements) using an LLM, with confidence scoring per test case (LLM self-assessment plus heuristics: assertion quality, naming, completeness), batching uncertain cases for human review, and integration with the team's existing test framework. Use when the user asks to generate unit tests from acceptance criteria, convert user stories to test cases, automate test creation from requirements, or augment a spec-driven test suite with AI-generated stubs that are then curated before merge.
日本語の概要は準備中です。原文の説明を表示しています。
Configures Allure Report (test-runner adapter install, `allure-results` directory wiring, `categories.json` for failure classification, `history-trend.json` retention via the copy-history-between-runs pattern), runs the Allure CLI to convert `allure-results` to a static HTML site, and uploads the report as a CI artifact. Use when the team needs richer test reporting than JUnit XML - step-level attachments, per-test history, retry tracking, and severity / epic / feature labeling across framework-agnostic adapters (pytest, Jest, JUnit, TestNG, NUnit, Mocha). As a rich static HTML report generator, it is the open-source alternative to the sunset ExtentReports (JVM/.NET per-test HTML narrative); for hosted cross-run flakiness analytics rather than a static per-run report use currents-integration.
日本語の概要は準備中です。原文の説明を表示しています。