[{"data":1,"prerenderedAt":119},["ShallowReactive",2],{"guide:automated-accessibility-checks-limits":3},{"slug":4,"title":5,"description":8,"keywords":11,"updatedAt":14,"summary":15,"sections":24,"relatedPages":74},"automated-accessibility-checks-limits",{"zh-Hant-TW":6,"en":7},"自動化無障礙檢測能找到什麼？又找不到什麼？","What Automated Accessibility Checks Can and Cannot Find",{"zh-Hant-TW":9,"en":10},"自動化檢測很適合找出缺少標籤、ARIA 錯誤、部分對比與結構問題，但仍需要人工檢查內容、流程、鍵盤操作與輔助科技情境。","Automated checks are useful for missing labels, ARIA errors, some contrast and structure issues, but manual review is still needed for content, flows, keyboard operation, and assistive technology context.",{"zh-Hant-TW":12,"en":13},"自動化無障礙檢測限制, axe-core 檢測, WCAG 自動化檢測, 人工無障礙檢測","automated accessibility testing limitations, axe-core testing, WCAG automated checks, manual accessibility review","2026-07-25",{"zh-Hant-TW":16,"en":20},[17,18,19],"自動化檢測是第一層篩選，不是最後結論。","它能快速找出一部分可被機器判斷的 WCAG 與最佳實務問題。","主要任務流程、內容脈絡、鍵盤操作與輔助科技使用經驗仍需要人工確認。",[21,22,23],"Automated checks are a first filter, not the final conclusion.","They quickly surface some machine-detectable WCAG and best-practice issues.","Core task flows, content context, keyboard use, and assistive technology experience still need manual review.",[25,36,47,65],{"title":26,"body":29},{"zh-Hant-TW":27,"en":28},"先用它找出明顯風險","Use it first to find obvious risks",{"zh-Hant-TW":30,"en":33},[31,32],"自動化檢測適合放在開發、QA、內容上架與 release 前流程。它可以快速指出缺少可及名稱、錯誤 ARIA、部分色彩對比、表單標籤與頁面結構問題。","這些結果應該變成待確認清單，而不是單純追求分數。",[34,35],"Automated checks fit well in development, QA, content publishing, and pre-release review. They can quickly identify missing accessible names, ARIA errors, some contrast issues, form labels, and page structure problems.","The result should become a review list, not just a score to optimize.",{"title":37,"body":40},{"zh-Hant-TW":38,"en":39},"再用人工檢查完成判斷","Then complete the judgment manually",{"zh-Hant-TW":41,"en":44},[42,43],"人工檢查應該至少走過主要任務流程：只用鍵盤操作、確認焦點順序、看錯誤訊息是否能讓人繼續、檢查圖片替代文字是否符合脈絡。","如果產品面向大量使用者，還應納入輔助科技測試與真實使用者回饋。",[45,46],"Manual review should cover key tasks: use only the keyboard, confirm focus order, check whether error messages help users continue, and review whether image alternatives fit the context.","For products used by many people, assistive technology testing and real user feedback should also be part of the process.",{"title":48,"body":51,"items":56},{"zh-Hant-TW":49,"en":50},"把兩者接成流程","Connect both into one workflow",{"zh-Hant-TW":52,"en":54},[53],"DevCheck 可以先在目前瀏覽器分頁執行機器掃描，再搭配人工檢視輔助、情境模擬與 AI 輔助建議做人工檢查。Pulse 則讓上線後仍發生的操作困難回到團隊視野。",[55],"DevCheck can run machine checks in the current browser tab, then support manual review through review aids, simulations, and AI-assisted suggestions. Pulse brings post-launch interaction barriers back into the team’s view.",{"zh-Hant-TW":57,"en":61},[58,59,60],"先掃描：找出可被機器偵測的風險。","再操作：用鍵盤、縮放、情境模擬與主要任務流程確認。","後追蹤：上線後觀察回報與互動困難是否仍重複出現。",[62,63,64],"Scan first: find machine-detectable risks.","Operate next: check keyboard, zoom, simulations, and key task flows.","Track later: observe reports and repeated interaction barriers after launch.",{"title":66,"body":69},{"zh-Hant-TW":67,"en":68},"哪些結果不應被自動化分數掩蓋","Which findings should not be hidden behind a score",{"zh-Hant-TW":70,"en":72},[71],"自動化工具常會把結果整理成數量或分數，但 release 判斷不能只看總分。少數幾個問題可能比大量低風險提醒更嚴重，特別是會阻擋主要任務、造成鍵盤使用者卡住，或讓使用者不知道如何修正錯誤的問題。",[73],"Automated tools often summarize results as counts or scores, but release decisions should not rely only on totals. A few issues can be more serious than many low-risk hints, especially when they block a main task, trap keyboard users, or prevent people from knowing how to fix an error.",[75,82,89,94,99,104,109,114],{"label":76,"path":78,"description":79},{"zh-Hant-TW":77,"en":77},"Accesserty DevCheck","\u002Fdevcheck",{"zh-Hant-TW":80,"en":81},"用瀏覽器內工具檢查本機、測試站、登入後頁面與互動狀態。","Run browser-based checks on local builds, staging pages, authenticated screens, and interactive states.",{"label":83,"path":85,"description":86},{"zh-Hant-TW":84,"en":84},"Accesserty Pulse","\u002Fpulse",{"zh-Hant-TW":87,"en":88},"觀察上線後的操作困難訊號、機器掃描摘要與使用者回報。","Observe post-launch interaction signals, machine scan summaries, and user reports.",{"label":90,"path":93},{"zh-Hant-TW":91,"en":92},"自動化與人工檢測術語頁","Automated vs manual testing glossary page","\u002Fglossary\u002Fautomated-vs-manual-accessibility-testing",{"label":95,"path":98},{"zh-Hant-TW":96,"en":97},"可及名稱實作指南","Accessible name implementation guide","\u002Fguides\u002Fhow-to-implement-accessible-names",{"label":100,"path":103},{"zh-Hant-TW":101,"en":102},"焦點路徑檢查指南","Focus path review guide","\u002Fguides\u002Fhow-to-check-keyboard-focus-path",{"label":105,"path":108},{"zh-Hant-TW":106,"en":107},"WCAG 檢測清單","WCAG testing checklist","\u002Fguides\u002Fwcag-testing-checklist",{"label":110,"path":113},{"zh-Hant-TW":111,"en":112},"圖片替代文字指南","Image alt text guide","\u002Fguides\u002Fwriting-meaningful-alt-text",{"label":115,"path":118},{"zh-Hant-TW":116,"en":117},"PDF 無障礙術語頁","PDF accessibility glossary page","\u002Fglossary\u002Fpdf-accessibility",1789828081462]