copilot-mcp-server
Access GitHub Copilot AI tools via MCP
This skill provides direct access to GitHub Copilot AI models through Model Context Protocol tools. Use it for code review, debugging, refactoring, and test generation without leaving your AI coding assistant.
Baixar o ZIP da skill
Upload no Claude
Vá em Configurações → Capacidades → Skills → Upload skill
Ative e comece a usar
Testar
A utilizar "copilot-mcp-server". Review this code for security issues: function login(u, p) { return db.query('SELECT * WHERE u=' + u + ' AND p=' + p); }
Resultado esperado:
- CRITICAL: SQL Injection Vulnerability - user input is directly concatenated into SQL query
- RECOMMENDATION: Use parameterized queries or an ORM
- HIGH: Password passed in plain text - hash and salt passwords
- MEDIUM: No rate limiting on login attempts
A utilizar "copilot-mcp-server". Debug this error: TypeError: Cannot read property 'map' of undefined
Resultado esperado:
- ROOT CAUSE: Data is undefined when attempting to call .map()
- SUGGESTION: Add null check before mapping: data?.map() or use optional chaining
- CONTEXT: Check if API response is being correctly assigned before rendering
A utilizar "copilot-mcp-server". Generate tests for a function that calculates fibonacci numbers
Resultado esperado:
- Test case 1: fibonacci(0) should return 0
- Test case 2: fibonacci(1) should return 1
- Test case 3: fibonacci(5) should return 5
- Test case 4: fibonacci(10) should return 55
- Test case 5: Negative input should throw error or return null
Auditoria de Segurança
SeguroThis is a documentation-only skill containing no executable code. The static findings are all false positives: model names and repository URLs were misidentified as cryptographic algorithms and C2 keywords; JavaScript code examples with parentheses were flagged as backtick execution; and documentation text was misinterpreted. This skill provides only markdown documentation and metadata about GitHub Copilot MCP server integration.
Fatores de risco
🌐 Acesso à rede (1)
⚙️ Comandos externos (28)
Pontuação de qualidade
O Que Você Pode Construir
Professional code review
Get detailed code reviews focusing on security, performance, and maintainability with specific focus areas
Error troubleshooting
Debug code errors with context-aware analysis and receive actionable fixes and explanations
Test generation
Generate unit tests for existing code using frameworks like Jest, pytest, or Mocha
Tente Estes Prompts
Ask Copilot how to implement a REST API endpoint for user authentication with JWT tokens
Review this code for security vulnerabilities and SQL injection risks: [paste code]
Debug this code with error: [error message]. Code: [paste code]
Generate Jest unit tests for this function: [paste code]
Melhores Práticas
- Be specific in prompts and provide relevant context
- Choose appropriate model based on task complexity
- Use session management for related queries to maintain context
Evitar
- Using overly broad prompts like 'write code' without specifics
- Overusing powerful models for simple tasks
- Mixing unrelated queries in a single session
Perguntas Frequentes
Which AI models are supported?
What is the request limit?
How do I configure the MCP server?
Is my code data safe?
What if MCP server is unavailable?
How is this different from copilot-flow?
Detalhes do Desenvolvedor
Autor
AykahshiLicença
MIT
Repositório
https://github.com/Aykahshi/copilot-mcp-tool/tree/main/plugins/copilot-flow/skills/copilot-mcp-serverReferência
main
Estrutura de arquivos
📄 SKILL.md