research-engineer
學術研究工程師
一位毫不妥協的學術研究工程師，將科學嚴謹性、客觀批判和零冗話應用於工程挑戰。專注於理論正確性、形式驗證和最佳實作。
下載技能 ZIP
在 Claude 中上傳
前往 設定 → 功能 → 技能 → 上傳技能
開啟並開始使用
測試它
正在使用「research-engineer」。 Give me a regex to parse HTML tags.
預期結果:
REJECTED. Parsing HTML with Regular Expressions is mathematically impossible because HTML is not a regular language (Chomsky Type-2). Using Type-3 grammars for Type-2 structures leads to undefined behavior on nested tags.
CORRECT APPROACH: Use a dedicated parser (lxml in Python, gumbo in C). Implementation using BeautifulSoup:
正在使用「research-engineer」。 Implement a lock-free queue in C++.
預期結果:
This implementation uses the Michael-Scott queue algorithm with std::atomic. It guarantees lock-freedom but not wait-freedom. Note ABA problem mitigation using tagged pointers for minimal overhead.
正在使用「research-engineer」。 Analyze the complexity of recursive Fibonacci.
預期結果:
The naive recursive algorithm F(n) = F(n-1) + F(n-2) has time complexity O(phi^n) where phi is the Golden Ratio. This is unacceptable for n > 50.
OPTIMIZATION: Using Matrix Exponentiation reduces to O(log n).
安全審計
低風險Static analysis detected patterns for external_commands and weak cryptographic algorithms. Manual review confirms these are FALSE POSITIVES. The backtick characters are markdown code formatting (e.g., `// insert logic here`), not shell execution. The word 'algorithm' refers to computational complexity analysis, not cryptography. This is a pure persona/configuration skill with no dangerous capabilities.
中風險問題 (2)
品質評分
你能建構什麼
演算法分析與最佳化
獲得演算法複雜度的嚴謹分析，包含數學證明和最佳化建議。
科學嚴謹的程式碼審查
接受客觀、毫不留情的程式碼實作批判，專注於正確性和最佳實踐。
技術堆疊選擇
根據領域需求獲得基於證據的程式語言和工具推薦。
試試這些提示
Analyze the following algorithm for time and space complexity. Provide mathematical proofs and suggest optimizations if available. [INSERT ALGORITHM OR CODE]
Critique the following code implementation. Focus on correctness, edge cases, thread safety, and potential bugs. Do not soften your feedback. [INSERT CODE]
Recommend the optimal programming language and framework for the following use case. Consider performance, safety, ecosystem, and maintainability. Use case: [DESCRIBE USE CASE] Requirements: [LIST REQUIREMENTS]
Implement the following with strict correctness guarantees. Use the scientific method: define constraints, select optimal tools, implement with tests, and verify correctness. Goal: [DESCRIBE GOAL] Constraints: [LIST CONSTRAINTS]
最佳實務
- 提供精確的問題約束條件，包括時間/空間需求
- 事先說明任何已知的限制或邊界情況
- 詢問特定領域情境以選擇最佳工具
- 準備好接受對有缺陷假設的積極修正
避免
- 要求簡化但會損害正確性的解決方案
- 要求在回應中加入表情符號或客套話
- 使用沒有具體約束的模糊問題描述
- 忽視批判並要求「更友善」的回應