# Code Review Prompt

Use this prompt when asking AI assistants to review code.

## Prompt Template

```
Please review this code following the standards in .ai/context/AGENTS_[LANGUAGE].md:

[PASTE CODE HERE]

Focus on:
1. **Standards compliance**: Does it follow patterns from AGENTS_*.md?
2. **Type safety**: Proper type annotations (Go, Python, TypeScript)?
3. **Error handling**: Are errors handled properly?
4. **Testing**: Are tests needed? Do they follow patterns?
5. **Documentation**: Is it properly documented?
6. **Performance**: Any obvious inefficiencies?
7. **Security**: Any security concerns?

For LazyVim users, suggest relevant keybindings from .ai/context/EDITORS.md.
```

## Language-Specific Reviews

### Go

```
Review this Go code against .ai/context/AGENTS_GO.md:

[CODE]

Check:
- Does it accept interfaces and return concrete types?
- Are functional options used for 3+ optional parameters?
- Is context passed properly?
- Are tests table-driven with subtests?
- Is error handling idiomatic?
```

### Python

```
Review this Python code against .ai/context/AGENTS_PYTHON.md:

[CODE]

Check:
- Are type hints complete (mypy strict mode)?
- Is it using uv for dependencies?
- Is Pydantic used for validation?
- Are async patterns correct?
- Do tests follow pytest conventions?
```

### TypeScript/Vue

```
Review this Vue component against .ai/context/AGENTS_TYPESCRIPT_VUE.md:

[CODE]

Check:
- Is it using Composition API with <script setup>?
- Are TypeScript types explicit (no 'any')?
- Are props/emits properly typed?
- Is it following Vue 3 best practices?
- Are composables used appropriately?
```

### PowerShell

```
Review this PowerShell script against .ai/context/AGENTS_POWERSHELL.md:

[CODE]

Check:
- Does it handle multi-subscription by default?
- Is it using Graph API instead of loops?
- Is Azure session validated?
- Are errors handled properly?
- Is it following approved verb conventions?
```

## Example Usage

```bash
# In Claude.ai or other AI assistant
"Please review this Go service following .ai/context/AGENTS_GO.md"

[paste code]
```

## Tips

- Reference specific sections: "Check against the 'Accept interfaces, return concrete' pattern"
- Ask for specific feedback: "Focus on error handling and testing"
- Request refactoring: "How would this look following our standards?"
- Get LazyVim help: "What keybindings would help me refactor this?"

## Output Format

Ask the AI to structure feedback as:

```markdown
## Compliance Issues

- [Issue description with line numbers]

## Suggestions

- [Improvement with example]

## Approved

- [What's already good]

## Refactored Code

[Improved version following standards]
```