# CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

## Commands

```bash
pnpm check              # Merge gate — must pass before every commit (audit + knip + biome + typecheck + test/coverage + jscpd)
pnpm knip               # Unused code detection (Knip)
pnpm test               # Run all tests (Vitest)
pnpm test:watch         # Tests in watch mode
pnpm typecheck          # tsc --build (project references)
pnpm lint               # Biome lint
pnpm format             # Biome format (writes)
pnpm --filter web dev   # Vite dev server (localhost:5173)
pnpm --filter web build # Production build
```

Run a single test file: `pnpm vitest run packages/domain/src/__tests__/advance-turn.test.ts`

## Architecture

Strict layered architecture with ports/adapters and enforced dependency direction:

```
apps/web (React 19 + Vite)  →  packages/application (use cases)  →  packages/domain (pure logic)
```

- **Domain** — Pure functions, no I/O, no framework imports. All state transitions are deterministic. Errors returned as values (`DomainError`), never thrown. Adapters may throw only for programmer errors.
- **Application** — Orchestrates domain calls via port interfaces (`EncounterStore`, `BestiarySourceCache`). No business logic here.
- **Web** — React adapter. Implements ports using hooks/state. All UI components, routing, and user interaction live here.

Layer boundaries are enforced by `scripts/check-layer-boundaries.mjs`, which runs as a Vitest test. Domain and application must never import from React, Vite, or upper layers.

### Data & Storage

- **localStorage** — encounter persistence (adapter layer, JSON serialization)
- **IndexedDB** — bestiary source cache (`apps/web/src/adapters/bestiary-cache.ts`, via `idb` wrapper)
- **`data/bestiary/index.json`** — pre-built search index for creature lookup, generated by `scripts/generate-bestiary-index.mjs`

### Project Structure

```
apps/web/                  React app — components, hooks, adapters
packages/domain/src/       Pure state transitions, types, validation
packages/application/src/  Use cases, port interfaces
data/bestiary/             Bestiary search index
scripts/                   Build tooling (layer checks, index generation)
specs/NNN-feature-name/    Feature specs (spec.md, plan.md, tasks.md)
.specify/                  Speckit config (templates, scripts, constitution)
docs/agents/               RPI skill artifacts (research reports, plans)
.claude/skills/            Agent skills (rpi-research, rpi-plan, rpi-implement)
```

## Tech Stack

- TypeScript 5.8 (strict mode, `verbatimModuleSyntax`)
- React 19, Vite 6, Tailwind CSS v4
- Lucide React (icons)
- `idb` (IndexedDB wrapper for bestiary cache)
- Biome 2.0 (formatting + linting), Knip (unused code), jscpd (copy-paste detection)
- Vitest (testing, v8 coverage), Lefthook (pre-commit hooks)

## Conventions

- **Biome 2.0** for formatting and linting (no Prettier, no ESLint). Tab indentation, 80-char lines. Imports are auto-organized alphabetically.
- **TypeScript strict mode** with `verbatimModuleSyntax`. Use `.js` extensions in relative imports when required by the repo's ESM settings (e.g., `./types.js`).
- **Branded types** for identity values (e.g., `CombatantId`). Prefer immutability/`readonly` where practical.
- **Domain events** are plain data objects with a `type` discriminant — no classes.
- **Tests** live in `packages/*/src/__tests__/*.test.ts`. Test pure functions directly; map acceptance scenarios and invariants from specs to individual `it()` blocks.
- **Feature specs** live in `specs/NNN-feature-name/` with spec.md (and optionally plan.md, tasks.md for new work). Specs describe features, not individual changes. The project constitution is at `.specify/memory/constitution.md`.
- **Quality gates** are enforced at pre-commit via Lefthook's `pnpm check` — the project's single earliest enforcement point. No gate may exist only as a CI step or manual process.

## Speckit Workflow

Speckit (`/speckit.*` skills) manages the spec-driven development pipeline. Specs are **living documents** that describe features, not individual changes.

### Issue-driven workflow
- `/write-issue` — create a well-structured Gitea issue via interactive interview
- `/integrate-issue <number>` — fetch an issue, route it to the right spec, and update the spec with the new/changed requirements. Then implement directly.
- `/sync-issue <number>` — push acceptance criteria from the spec back to the Gitea issue

### RPI skills (Research → Plan → Implement)
- `rpi-research` — deep codebase research producing a written report in `docs/agents/research/`
- `rpi-plan` — interactive phased implementation plan in `docs/agents/plans/`
- `rpi-implement` — execute a plan file phase by phase with automated + manual verification

### Choosing the right workflow by scope

| Scope | Workflow |
|---|---|
| Bug fix / CSS tweak | Just fix it, commit |
| Small change to existing feature | `/integrate-issue` → implement → commit |
| Larger addition to existing feature | `/integrate-issue` → `rpi-research` → `rpi-plan` → `rpi-implement` |
| New feature | `/speckit.specify` → `/speckit.clarify` → `/speckit.plan` → `/speckit.tasks` → `/speckit.implement` |

Speckit manages **what** to build (specs as living documents). RPI manages **how** to build it (research, planning, execution). The full speckit pipeline is for new features. For changes to existing features, update the spec via `/integrate-issue`, then use RPI skills if the change is non-trivial.

### Current feature specs
- `specs/001-combatant-management/` — CRUD, persistence, clear, batch add, confirm buttons
- `specs/002-turn-tracking/` — rounds, turn order, advance/retreat, top bar
- `specs/003-combatant-state/` — HP, AC, conditions, concentration, initiative
- `specs/004-bestiary/` — search index, stat blocks, source management, panel UX

## Constitution (key principles)

The constitution (`.specify/memory/constitution.md`) governs all feature work:

1. **Deterministic Domain Core** — Pure state transitions only; no I/O, randomness, or clocks in domain.
2. **Layered Architecture** — Domain → Application → Adapters. Never skip layers or reverse dependencies.
3. **Clarification-First** — Ask before making non-trivial assumptions.
4. **MVP Baseline** — Say "MVP baseline does not include X", never permanent bans.
5. **Spec-driven features** — Features are described in living specs; evolve existing specs via `/integrate-issue`, create new ones via `/speckit.specify`. Bug fixes and tooling changes do not require specs.