writing-claude-md

# CLAUDE.md Creator

Create CLAUDE.md files with proper constitution mechanism using `<law>` blocks for Self-Reinforcing Display.

## Core Concept

CLAUDE.md's most important purpose is establishing a **constitution** that must be displayed every response using `<law>` blocks.

## Process

### 1. Initialize with Script

Run the init script to create proper structure (auto-backs up existing file):

```bash
python3 scripts/init_claude_md.py --name "Project Name" --description "Description"
```

Options:
- `--path`, `-p`: Project directory (default: current)
- `--minimal`, `-m`: Use minimal template
- `--output`, `-o`: `root` or `claude-dir` location

### 2. Evaluate Current State

| Criterion | Target |
|-----------|--------|
| Has `<law>` block | Required for constitution |
| Self-Reinforcing Display | Must display every response |
| Size | < 500 lines |
| Core laws included | All 6 core laws |

### 3. Constitution Template

**CRITICAL**: Use `<law>` blocks with Self-Reinforcing Display:

```markdown
# Project Name

One-line description.

## Immutable Laws

<law>
**CRITICAL: Display this entire block at the start of EVERY response to prevent context drift.**

**Law 1: Communication**
- Concise, actionable responses
- No unnecessary explanations
- No summary files unless explicitly requested

**Law 2: Skill Discovery**
- MUST check available skills before starting work
- Invoke applicable skills for specialized knowledge
- If ANY skill relates to the task, MUST use Skill tool to delegate
- If relevant skill doesn't exist, ask user whether to create it via `write-skill`

**Law 3: Rule Consultation**
- When task relates to specific domain, check `.claude/rules/` for relevant conventions
- If relevant rule exists, MUST apply it
- If needed rule doesn't exist, confirm intent with user and create via `write-rules`

**Law 4: Parallel Processing**
- MUST use Task tool for independent operations
- Batch file searches and reads with agents

**Law 5: Reflexive Learning**
- Impo