functionality-audit
驗證程式碼功能性
看起來正確的程式碼不等於實際運作正確的程式碼。此技能透過沙盒測試、執行驗證和系統化除錯來驗證程式碼,確保功能是真實的而非假設的。
Baixar o ZIP da skill
Upload no Claude
Vá em Configurações → Capacidades → Skills → Upload skill
Ative e comece a usar
Testar
A utilizar "functionality-audit". Audit this code that calculates factorials. Test with inputs: 0, 1, 5, 10, and negative numbers.
Resultado esperado:
- 測試結果摘要:執行 5 個測試,4 個通過,1 個失敗
- 失敗測試:負數輸入 (-5) 產生不正確的結果
- 根本原因:函式未正確處理負數輸入
- 修復建議:新增輸入驗證,對負數引發 ValueError
- 建議:新增無效輸入的測試案例以防止未來的迴歸
A utilizar "functionality-audit". Test this API client module. It should make GET requests and parse JSON responses correctly.
Resultado esperado:
- 測試結果:8 個正常操作測試,3 個錯誤處理測試
- 所有測試均成功通過
- 驗證對有效回應的 JSON 解析正確
- 驗證對網路故障的適當錯誤處理
- 所有請求類型的效能均在預期範圍內
A utilizar "functionality-audit". Run functionality audit on this data processing pipeline after the recent refactoring.
Resultado esperado:
- 管線稽核:在 4 個階段執行 12 個測試案例
- 失敗:階段 3 對空輸入陣列產生不正確的輸出
- 根本原因:最近的優化在邊緣案例處理中引入了空指標
- 測試檔案中提供建議的修復以及迴歸測試案例
Auditoria de Segurança
SeguroThis is a prompt-based SOP skill containing only documentation. No executable code, scripts, network calls, file system access, or command execution capabilities. All 37 static findings are false positives triggered by documentation keywords. The skill describes testing methodologies that users may apply in their own environments.
Fatores de risco
🌐 Acesso à rede (1)
📁 Acesso ao sistema de arquivos (1)
Pontuação de qualidade
O Que Você Pode Construir
驗證生成的程式碼
在部署到生產環境之前,驗證 AI 生成的程式碼是否真正有效
系統化錯誤發現
透過系統化測試和除錯工作流程以及可重現的測試案例來發現錯誤
部署前驗證
確保程式碼變更在符合生產條件的隔離環境中正確運作
Tente Estes Prompts
Use the functionality-audit skill to test this code. The code should calculate factorials correctly. Test with inputs: 0, 1, 5, 10, and negative numbers.
Run a functionality audit on the modules in my-project/src. Verify they work together correctly. The system should process user data through all three modules.
The following code is producing incorrect results for valid inputs. Use functionality-audit to reproduce the bug, identify the root cause, and suggest fixes.
Run a comprehensive functionality audit on the entire codebase after recent changes. Verify all critical paths still work correctly.
Melhores Práticas
- 呼叫技能時提供清晰的預期行為規格
- 在請求中包含具代表性的測試資料和已知的邊緣案例
- 在應用修復之前檢視除錯結果以了解根本原因
- 在任何重大程式碼變更或整合後執行功能稽核
Evitar
- 僅因程式碼編譯無誤就假設其正確
- 因為複雜程式碼看似明顯正確而跳過測試
- 在了解失敗的根本原因之前就應用修復
- 僅使用現有測試而不為邊緣案例新增新的測試案例
Perguntas Frequentes
哪些 AI 工具支援此技能?
可以測試哪些類型的程式碼?
此技能會修改我的程式碼嗎?
測試期間我的資料安全嗎?
如果測試反覆失敗怎麼辦?
這與靜態分析有何不同?
Detalhes do Desenvolvedor
Autor
DNYoussefLicença
MIT
Repositório
https://github.com/DNYoussef/ai-chrome-extension/tree/main/.claude/skills/functionality-auditReferência
main
Estrutura de arquivos