# Project Rules

## General
- Follow the existing code style and patterns.
- Use pnpm for running project commands.
- Keep code in TypeScript unless migration is required.
- When refactoring or creating new components, review `docs/frontend-layering.md` first and follow its layering and responsibility guidelines.
- When a change affects LLM editing boundaries, page creation flow, layout usage, login-page boundaries, or frontend layering rules, update `docs/llm-development-guide.md` in the same change.
- For UI debugging that spans implementation, state flow, and live browser behavior, use subagents deliberately: one for component contracts and event flow, one for data sources / routing / integration, and one for live browser verification. Do not edit files until the evidence from those threads converges on a root cause.
- Treat subagent output as scoped evidence, not as a final answer. Use subagents to narrow the search space, confirm or eliminate hypotheses, and reduce local context load before making edits.
- When a problem is directly tied to Vuetify component behavior, props, slots, accessibility output, or generated DOM, consult Vuetify MCP and official Vuetify API documentation before changing the implementation. Prefer supported Vuetify props, slots, and documented extension points over custom replacements unless the documented API is insufficient.

## Stack
- Framework: Vue 3 + Vite
- UI Library: Vuetify
- Enabled Features: Vuetify MCP, Pinia, Vue I18n, Vue Router

## Icon Usage
```js
<script setup>
import { mdiAccount } from '@mdi/js'
</script>

<template>
  <v-icon :icon="mdiAccount" />
</template>
```

<!-- code-review-graph MCP tools -->
## MCP Tools: code-review-graph

**IMPORTANT: This project has a knowledge graph. ALWAYS use the
code-review-graph MCP tools BEFORE using Grep/Glob/Read to explore
the codebase.** The graph is faster, cheaper (fewer tokens), and gives
you structural context (callers, dependents, test coverage) that file
scanning cannot.

### When to use graph tools FIRST

- **Exploring code**: `semantic_search_nodes` or `query_graph` instead of Grep
- **Understanding impact**: `get_impact_radius` instead of manually tracing imports
- **Code review**: `detect_changes` + `get_review_context` instead of reading entire files
- **Finding relationships**: `query_graph` with callers_of/callees_of/imports_of/tests_for
- **Architecture questions**: `get_architecture_overview` + `list_communities`

Fall back to Grep/Glob/Read **only** when the graph doesn't cover what you need.

### Key Tools

| Tool | Use when |
|------|----------|
| `detect_changes` | Reviewing code changes — gives risk-scored analysis |
| `get_review_context` | Need source snippets for review — token-efficient |
| `get_impact_radius` | Understanding blast radius of a change |
| `get_affected_flows` | Finding which execution paths are impacted |
| `query_graph` | Tracing callers, callees, imports, tests, dependencies |
| `semantic_search_nodes` | Finding functions/classes by name or keyword |
| `get_architecture_overview` | Understanding high-level codebase structure |
| `refactor_tool` | Planning renames, finding dead code |

### Workflow

1. The graph auto-updates on file changes (via hooks).
2. Use `detect_changes` for code review.
3. Use `get_affected_flows` to understand impact.
4. Use `query_graph` pattern="tests_for" to check coverage.