ai_test_process_skills

Create a frozen HTML test execution report from current Japanese test artifacts and latest execution result folders. Use when Codex needs to snapshot test plans, analysis, design, test cases, review results, questionnaires, executed results, bug lists, unanswered/question-wait cases, unimplemented cases, source freshness metadata, and raw evidence into `テスト成果物/report/yyyyMMddHHmmss_report/`.

Create Japanese Markdown test analysis artifacts from a test plan, specifications, README files, existing tests, and implementation notes. Use when Codex needs to run a three-pass analysis that first derives specification-driven test viewpoints (テスト観点), then adds generally expected QA viewpoints for the target type, then re-examines whether additional missing viewpoints should be added, preserves traceability, updates the test plan when new risks or approaches should be added, and saves a Markdown questionnaire for open questions.

Create or refine Japanese Markdown test plan documents from product requirements, README files, specifications, existing tests, implementation notes, and user-provided source data. Use when Codex needs to draft a new test plan (テスト計画書) or re-run the test plan creation skill to further consider and add overlooked product risks and traceable test approaches. The plan defines test objectives, referenced documents, test items, scope, exit criteria, product risk scoring, traceable high-level test approaches, use-case/scenario testing strategy, formal/regression test asset handling, unresolved assumptions, and downstream analysis/design depth expectations.

Orchestrate the full Japanese QA workflow by invoking existing skills in order, from test plan creation through reviews, analysis, design, test cases, code-based and Playwright E2E test implementation, execution recording, final HTML report generation, and final report review. Use when Codex needs to run the whole test artifact lifecycle with provided specifications and test targets while leaving human-executed tests unexecuted and visible in the final report. Runs as a single user-visible workflow while using checkpointed phase handoffs to keep long sessions reliable.

Create Japanese Markdown test case tables from test design artifacts, design questionnaires, test analysis, test plans, specifications, README files, existing tests, and implementation notes. Use when Codex needs to expand each TDxxx design pattern into execution-ready, independently reportable TC-* test cases, output separate files for code-based tests, E2E automated tests, human-executed tests, question-wait cases, and a companion CSV for human-executed cases, and preserve coverage from test design to test cases.

Create Japanese Markdown test design artifacts from test analysis viewpoints, questionnaires, test plans, specifications, README files, existing tests, and implementation notes. Use when Codex needs to expand test viewpoints into traceable, sufficiently decomposed test designs (テスト設計), consume source-listed boundaries and finite value sets, define execution methods and expected-result judgment points, and create a Markdown questionnaire for unknowns from テスト分析.

Review Japanese Markdown test analysis artifacts and questionnaires against a test plan, product specifications, and QA analysis quality criteria. Use when Codex needs to assess or improve テスト分析, including both missing viewpoints and whether the viewpoints are necessary and sufficient for the product risks, prioritize findings, fix P0/P1/P2 fix-worthy issues in the analysis artifacts, re-review until no P0/P1/P2 fix-worthy issues remain, and save the final review result as Markdown.

Review Japanese QA artifacts end-to-end across test plans, test analysis, questionnaires, test design, and test case files. Use when Codex needs to find gaps, missing coverage, weak traceability, contradictions, or insufficient detail across テスト計画書, テスト分析, テスト設計, and テストケース; prioritize findings, fix high-priority issues across the appropriate QA artifacts, re-review until no high-priority fix-worthy issues remain, and save the final cross-artifact review result as Markdown.

Review Japanese Markdown and CSV test case artifacts against test design, questionnaires, test analysis, test plans, specifications, and QA execution quality criteria. Use when Codex needs to assess or improve テストケース files, including separate code-based, E2E automated, human-executed, question-wait Markdown files, the companion human-executed CSV, traceability from TC-* to TDxxx/TVxxx/TAxxx, question-wait handling, one-TC-one-independently-reportable-judgment readiness, or test-case executability; prioritize findings, fix P0/P1/P2 fix-worthy issues, re-review until no P0/P1/P2 fix-worthy issues remain, and save the final review result as Markdown.

Review Japanese Markdown test design artifacts and design questionnaires against test analysis, test plans, specifications, and QA design quality criteria. Use when Codex needs to assess or improve テスト設計, traceable test design tables, necessary-and-sufficient design depth, source-listed boundary/value coverage, test level/type grouping, execution-method clarity, or design questionnaires; prioritize findings, fix P0/P1/P2 fix-worthy issues, re-review until no P0/P1/P2 fix-worthy issues remain, and save the final review result as Markdown.

Review Japanese HTML/Markdown test execution reports against QA artifacts, execution results, issue lists, review results, and evidence completeness. Use when Codex needs to assess or improve a final テスト実行レポート, report snapshot folder, index.html, summary counts, Fail/issue visibility, not-run/question-wait/unimplemented visibility, artifact links, raw evidence links, source freshness, traceability, security redaction, or report readability; prioritize findings, fix P0/P1/P2 fix-worthy issues, re-review until no P0/P1/P2 fix-worthy issues remain, and save the final review result as Markdown.

Review Japanese Markdown test plans against product specifications and QA planning quality criteria, prioritize findings, fix high-priority issues in the test plan, re-review until no high-priority fix-worthy issues remain, and save the final review result as Markdown. Use when Codex needs to assess or improve a テスト計画書, test strategy document, QA plan, traceable test planning artifact, use-case/scenario testing strategy, generic QA coverage, formal/regression test asset handling, unresolved assumptions, or downstream analysis/design depth expectations.

Review Playwright E2E automated test implementations against traceable Markdown E2E test cases, implementation reports, unimplemented-case lists, target pages, and project configuration. Use when Codex needs to assess or improve Playwright tests generated from `テストケース_E2E自動.md` / `TC-E2E-*`; prioritize findings, fix P0/P1/P2 fix-worthy issues in test code or related QA artifacts, rerun or statically validate, repeat until no P0/P1/P2 fix-worthy issues remain, and save the final review result as Markdown.

Review executable code-based test implementations against traceable Markdown test cases, implementation reports, unimplemented-case lists, specifications, and target code. Use when Codex needs to assess or improve unit, integration, regression, or code-level automated tests generated from `テストケース_コードベース.md` / `TC-CB-*`; prioritize findings, fix P0/P1/P2 fix-worthy issues in test code or related QA artifacts, rerun tests, repeat until no P0/P1/P2 fix-worthy issues remain, and save the final review result as Markdown.

Create Playwright E2E automated test code from traceable Markdown E2E test cases. Use when Codex needs to turn `テストケース_E2E自動.md` / `TC-E2E-*` rows into JavaScript Playwright tests for a user-specified target file, URL, app entrypoint, component, or page; preserve TC/TD/TV/TA/R traceability, create minimal Playwright configuration when absent, and save question-wait or unsupported cases to `未実装テストケース_E2E自動.md`.

Create executable test code from traceable code-based Markdown test cases and separately document unimplemented code-based cases. Use when Codex needs to turn `テストケース_コードベース.md`, `TC-CB-*` rows, or similar code-level test case tables into actual unit, integration, regression, or code-level automated tests for a user-specified target file, module, class, function, or component; choose the appropriate language/framework from the repository, preserve TC/TD/TV/TA/R traceability, save question-wait or `要確認` cases to `未実装テストケース_コードベース.md`, and stop when no concrete test target is provided.

Execute code-based automated tests and record traceable run results. Use when Codex needs to run tests derived from `テストケース_コードベース.md`, create a timestamped report folder under `テスト成果物/report`, save executed Markdown/CSV/raw log with Pass/Fail/N/A per `TC-CB-*`, and write discovered bugs to that run's `コードベース_発見issue一覧.md` with `BUG-001` style IDs.

Execute Playwright E2E automated tests and record traceable run results. Use when Codex needs to run tests derived from `テストケース_E2E自動.md`, parse Playwright JSON reporter output for `TC-E2E-*`, create a timestamped report folder under `テスト成果物/report`, save executed Markdown/CSV/raw log/JSON with Pass/Fail/N/A, and write discovered bugs to that run's `e2e_発見issue一覧.md` with `E2E-BUG-001` style IDs.