QA Trace - AI Bug Reports, Private & Local 作成者: Anastasia Nesterova
Turn an exploratory testing session into an AI-ready bug report with steps to reproduce. Private and local.
1 人のユーザー1 人のユーザー
拡張機能メタデータ
スクリーンショット
この拡張機能について
QA Trace turns an exploratory testing session into a clear, reproducible, AI-ready bug report - without sending your data to anyone's cloud.
Track what you click and type, capture console, network, and UI errors automatically, then generate a report-ready prompt for steps to reproduce, test documentation, or a full session report. Paste it into any AI (ChatGPT, Claude, DeepSeek) or your own endpoint - or copy it as plain text.
★ PRIVATE BY DESIGN
- No account, no sign-up, no tracking.
- Your data stays in your browser and auto-deletes after 12 hours.
- Runs only on the exact sites you allow.
- Sensitive headers (Authorization, Cookie, API keys) and body fields (passwords, tokens) are redacted automatically.
- The site's origin is stripped before anything leaves your browser. Nothing is uploaded unless you turn on an optional integration yourself.
=> Usable on internal, staging, and enterprise environments where cloud bug-reporters are banned.
★ BUILT FOR REPRODUCTION
- Captures clicks, double-clicks, input, changes, tab open/reload (passwords masked).
- Detects console errors, failed fetch/XHR requests (status, headers, body - including a 5xx hidden inside a 200 response), and UI errors by CSS selector with an automatic screenshot.
- Optional full fetch/XHR logging per site; copy any request as JSON or download the whole log.
★ AI-READY OUTPUT, YOUR MODEL
- 3 modes: Steps to Reproduce for a ticket, neutral Document Steps, or a Full Report of actions + errors.
- Bring your own LLM: OpenAI, DeepSeek, or any OpenAI-compatible endpoint - or just copy the prompt.
- Bilingual prompt templates: English and Russian.
★ FLEXIBLE WORKFLOW
- Scope by tab, action count, or time window.
- Mark expected errors to exclude them.
- Describe unexpected behavior in free text.
- Send to your LLM, trigger a webhook, or copy to clipboard.
★ FREE & CROSS-BROWSER
- Completely free, no seats, no paywall.
- Works on Chrome, Firefox, and Yandex Browser.
WHY THE BROAD HOST PERMISSION?
QA teams test on many different internal and staging origins that can't be known in advance, so the content script is available everywhere, but tracking activates ONLY on origins you explicitly add to your Allowed URLs list. On every other page it checks the list, finds no match, and does nothing.
Made by a QA engineer, for QA engineers.
Track what you click and type, capture console, network, and UI errors automatically, then generate a report-ready prompt for steps to reproduce, test documentation, or a full session report. Paste it into any AI (ChatGPT, Claude, DeepSeek) or your own endpoint - or copy it as plain text.
★ PRIVATE BY DESIGN
- No account, no sign-up, no tracking.
- Your data stays in your browser and auto-deletes after 12 hours.
- Runs only on the exact sites you allow.
- Sensitive headers (Authorization, Cookie, API keys) and body fields (passwords, tokens) are redacted automatically.
- The site's origin is stripped before anything leaves your browser. Nothing is uploaded unless you turn on an optional integration yourself.
=> Usable on internal, staging, and enterprise environments where cloud bug-reporters are banned.
★ BUILT FOR REPRODUCTION
- Captures clicks, double-clicks, input, changes, tab open/reload (passwords masked).
- Detects console errors, failed fetch/XHR requests (status, headers, body - including a 5xx hidden inside a 200 response), and UI errors by CSS selector with an automatic screenshot.
- Optional full fetch/XHR logging per site; copy any request as JSON or download the whole log.
★ AI-READY OUTPUT, YOUR MODEL
- 3 modes: Steps to Reproduce for a ticket, neutral Document Steps, or a Full Report of actions + errors.
- Bring your own LLM: OpenAI, DeepSeek, or any OpenAI-compatible endpoint - or just copy the prompt.
- Bilingual prompt templates: English and Russian.
★ FLEXIBLE WORKFLOW
- Scope by tab, action count, or time window.
- Mark expected errors to exclude them.
- Describe unexpected behavior in free text.
- Send to your LLM, trigger a webhook, or copy to clipboard.
★ FREE & CROSS-BROWSER
- Completely free, no seats, no paywall.
- Works on Chrome, Firefox, and Yandex Browser.
WHY THE BROAD HOST PERMISSION?
QA teams test on many different internal and staging origins that can't be known in advance, so the content script is available everywhere, but tracking activates ONLY on origins you explicitly add to your Allowed URLs list. On every other page it checks the list, finds no match, and does nothing.
Made by a QA engineer, for QA engineers.
0 人のレビュー担当者が 0 と評価しました
権限とデータ
必要な権限:
- クリップボードへのデータ入力
- 通知の表示
- ブラウザーのタブへのアクセス
- ナビゲーション中のブラウザーアクティビティへのアクセス
- すべてのウェブサイトの保存されたデータへのアクセス
データ収集:
- 開発者によると、この拡張機能はデータ収集を必要としません。
詳しい情報
- アドオンリンク
- バージョン
- 1.1.1
- サイズ
- 116.31 KB
- 最終更新日
- 23日前 (2026年9月6日)
- 関連カテゴリー
- プライバシーポリシー
- このアドオンのプライバシーポリシーを読む
- バージョン履歴
- コレクションへ追加