Web applications can accumulate insecure settings, deprecated APIs, and browser compatibility problems. This skill guides evidence-based reviews and targeted fixes using Lighthouse, source inspection, and security checks.

Slow loading, delayed interactions, and unexpected layout shifts damage page experience. This skill guides evidence-based diagnosis and targeted fixes for LCP, INP, and CLS.

Local changes can hide bugs, security gaps, and missing tests before human review. This skill guides an evidence-based review with prioritized findings and explicit verification results.

Speculative fixes can hide causes and leave bugs unresolved. This skill guides assistants through reproduction, hypothesis testing, focused changes, and evidence-based verification.

Passing claims can hide skipped tests, stale results, or checks that never exercised a change. This skill guides fresh verification and evidence-based completion reports.

Reviewers need clear reasons, behavior changes, and testing evidence before assessing a pull request. This skill turns diffs and commits into structured descriptions with review priorities and rollout risks.

Redundant assertions can inflate confidence without protecting observable behavior. Captain Obvious combines deterministic scanners with contextual review and a test-authoring contract checklist.

Coding session logs make tool sequences and delays difficult to review. Agent Transit turns one local session into an interactive map with timing and failure markers.

Chrome extension development requires coordinating browser permissions, background behavior, user interfaces, and store requirements. This skill provides Manifest V3 examples and publishing templates to guide implementation and release preparation.

Media workflows often need reliable conversion, resizing, compression, and timing adjustments across formats. This skill provides practical FFmpeg commands for preparing media assets and Remotion inputs.

程式碼審查可能會遺漏品質、安全性與可維護性問題。此技能會透過腳本、檢查清單與報告產生流程，引導進行結構化審查。

混亂的 commit 歷史會讓審查與回滾更加困難。此 skill 會引導 Claude、Codex 或 Claude Code 完成 commit 策略、訊息撰寫與驗證。

開發者在加入本機模型功能之前，需要針對 Chrome 內建 AI 取得清楚指引。此技能說明 Prompt API 設定、工作階段、串流、結構化輸出、擴充功能與限制。

命令面板常因焦點錯誤、過期的監聽器，以及不穩定的命令參照而失效。這項技能提供實用的 React 模式，用於建立可靠的鍵盤驅動命令面板。

GitHub 登入與儲存庫存取通常需要分開的 OAuth 流程。此技能展示一種 Nango 雙連線模式，包含 webhook、資料庫對應和同步處理。

WebSocket 訊框錯誤很難追蹤，因為 HTTP upgrade 衝突可能看起來像是損壞的訊框。此 skill 會引導你診斷 Node.js ws 伺服器中的 RSV1、invalid header 與 code 1006 問題。

從 Cursor 遷移或撰寫有範圍限定的指引時，Claude 規則檔很容易設定錯誤。這項技能提供正確的 paths frontmatter、glob 範例，以及檔案組織指引。

Claude Code hooks 功能強大，但很容易格式設定錯誤。此 skill 為 Codex 和 Claude 提供一份精簡指南，涵蓋 hook 檔案、事件、驗證和發布步驟。

建立可靠的 Claude Code agents 需要正確的 frontmatter、工具限制，以及清楚的人設。這項 skill 會引導 Claude、Codex 和 Claude Code 完成 agent 結構、驗證與審查。

僅靠記錄檔很難確認視覺 UI 錯誤。此技能會引導使用瀏覽器驅動的檢查，包含截圖、元素選取與頁面檢查。