Habilidades research-engineer
📦

research-engineer

Riesgo bajo

Academic Research Engineer

An uncompromising Academic Research Engineer that applies scientific rigor, objective criticism, and zero fluff to engineering challenges. Focuses on theoretical correctness, formal verification, and optimal implementation.

Soporta: Claude Codex Code(CC)
📊 71 Adecuado
1

Descargar el ZIP de la skill

2

Subir en Claude

Ve a Configuración → Capacidades → Skills → Subir skill

3

Activa y empieza a usar

Pruébalo

Usando "research-engineer". Give me a regex to parse HTML tags.

Resultado esperado:

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:

Usando "research-engineer". Implement a lock-free queue in C++.

Resultado esperado:

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.

Usando "research-engineer". Analyze the complexity of recursive Fibonacci.

Resultado esperado:

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).

Auditoría de seguridad

Riesgo bajo
v1 • 2/24/2026

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.

1
Archivos escaneados
141
Líneas analizadas
2
hallazgos
1
Auditorías totales
Problemas de riesgo medio (2)
SKILL.md:28SKILL.md:40SKILL.md:78-79SKILL.md:92SKILL.md:99-101
False Positive: Markdown Code Formatting
Static scanner detected backtick characters as potential shell execution. These are actually markdown code formatting in a plain text skill file. The skill contains no executable code, shell commands, or cryptographic algorithms.
SKILL.md:3SKILL.md:136
False Positive: Algorithm Keyword
Static scanner flagged 'weak cryptographic algorithm' based on the word 'algorithm' appearing in the skill. Manual review shows this refers to computational complexity (e.g., 'algorithmic optimization'), not cryptography.
Auditado por: claude

Puntuación de calidad

38
Arquitectura
100
Mantenibilidad
87
Contenido
50
Comunidad
80
Seguridad
100
Cumplimiento de la especificación

Lo que puedes crear

Algorithm Analysis and Optimization

Get rigorous analysis of algorithm complexity with mathematical proofs and optimization recommendations.

Code Review with Scientific Rigor

Receive objective, unflinching critique of code implementations with focus on correctness and best practices.

Technology Stack Selection

Get evidence-based recommendations for programming languages and tools based on domain requirements.

Prueba estos prompts

Algorithm Review 
Analyze the following algorithm for time and space complexity. Provide mathematical proofs and suggest optimizations if available.

[INSERT ALGORITHM OR CODE]
Code Critique 
Critique the following code implementation. Focus on correctness, edge cases, thread safety, and potential bugs. Do not soften your feedback.

[INSERT CODE]
Technology Selection 
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]
Research Implementation 
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]

Mejores prácticas

  • Provide precise problem constraints including time/space requirements
  • State any known limitations or edge cases upfront
  • Ask for the specific domain context to select optimal tools
  • Be prepared for aggressive correction of flawed assumptions

Evitar

  • Asking for simplified solutions that compromise correctness
  • Requesting emoji or pleasantries in responses
  • Using vague problem descriptions without specific constraints
  • Ignoring the critique and asking for 'nicer' responses

Preguntas frecuentes

Why is this skill so harsh?
This skill prioritizes technical correctness over user feelings. In research and engineering, incorrect solutions have real consequences.
Can I get softer feedback?
No. The skill is designed for rigorous analysis. If you need gentler feedback, use a different skill.
What languages does this skill recommend?
It recommends based on domain: Fortran/Julia for numerical computing, Rust/Ada for safety-critical, Go/Rust for distributed systems, Coq/Lean for formal verification.
Does this skill write full implementations?
Yes, but only if required for correctness. It will not use placeholders or simplified stubs.
How does this skill handle impossible problems?
It immediately states if a problem is mathematically impossible or computationally intractable (e.g., NP-hard without approximation).
Can I use this for production code?
Yes, but expect rigorous scrutiny. The skill focuses on correctness and will highlight any shortcuts that compromise reliability.

Detalles del desarrollador

Autor

sickn33

Licencia

MIT

Repositorio

https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/research-engineer

Ref.

main

Estructura de archivos

📄 SKILL.md