Analyze / GitHub Copilot

Analyze Your Copilot Instructions

GitHub Copilot custom instructions shape how Copilot writes and reviews code. See what your instructions actually prioritize.

Repository-wide: .github/copilot-instructions.md — applies to all Copilot interactions in the repo.

Path-specific: .github/instructions/**/*.instructions.md — scoped rules with applyTo frontmatter for file pattern matching.

Tip: Copilot also reads AGENTS.md if present. Short, self-contained statements work best.

# Copilot Instructions

## General
- Write TypeScript with strict mode enabled
- Use functional components with hooks, not class components
- Prefer composition over inheritance
- Use descriptive variable names that reveal intent

## Error Handling
- Always handle promise rejections explicitly
- Use custom error classes for domain-specific errors
- Log errors with structured context (user, action, resource)
- Never catch and ignore errors silently

## API Design
- Use RESTful conventions for endpoint naming
- Validate all request inputs at the boundary
- Return consistent error response shapes
- Include pagination for list endpoints

## Security
- Sanitize all user input before database queries
- Use parameterized queries — never string concatenation
- Validate JWT tokens on every authenticated request
- Never expose internal error details to clients

## Testing
- Write unit tests for business logic
- Write integration tests for API endpoints
- Use fixtures and factories, not raw test data

Supports: Claude Code, OpenClaw, Cursor, GitHub Copilot, CrewAI, Windsurf, Aider, and more

Beyond Static Instructions

Copilot instructions are static — you write them once and update manually. Persona makes them living documents.

Import your instructions → connect your repos → discover patterns from actual code → export better instructions.

Also supported: CLAUDE.md · Cursor Rules · AGENTS.md · CrewAI · Windsurf · System Prompts