/new-project

<objective>

Initialize a new DOG project through comprehensive context gathering.

This is the most leveraged moment in any project. Deep questioning here means better task decomposition, better execution, better outcomes.

Creates .claude/dog/ with PROJECT.md, ROADMAP.md, and dog-config.json.

</objective>

<execution_context>

@${CLAUDE_PLUGIN_ROOT}/.claude/dog/templates/project.md @${CLAUDE_PLUGIN_ROOT}/.claude/dog/templates/roadmap.md @${CLAUDE_PLUGIN_ROOT}/.claude/dog/templates/dog-config.json @${CLAUDE_PLUGIN_ROOT}/.claude/dog/workflows/questioning.md

</execution_context>

<process> <step name="setup">

MANDATORY FIRST STEP - Execute these checks before ANY user interaction:

1. Abort if project exists:

   [ -f .claude/dog/PROJECT.md ] && echo "ERROR: Project already initialized. Use /dog:status" && exit 1
   

2. Initialize git repo in THIS directory (required even if inside a parent repo):

   if [ -d .git ] || [ -f .git ]; then
       echo "Git repo exists in current directory"
   else
       git init
       echo "Initialized new git repo"
   fi
   

3. Create planning directories:

   mkdir -p .claude/dog
   

   You MUST run all bash commands above using the Bash tool before proceeding.

</step> <step name="question">

1. Open (FREEFORM - do NOT use AskUserQuestion):

Ask inline: "What do you want to build?"

Wait for their freeform response. This gives you the context needed to ask intelligent follow-up questions.

2. Follow the thread (NOW use AskUserQuestion):

Based on their response, use AskUserQuestion with options that probe what they mentioned:

• header: "[Topic they mentioned]"
• question: "You mentioned [X] - what would that look like?"
• options: 2-3 interpretations + "Something else"

3. Sharpen the core:

Use AskUserQuestion:

• header: "Core"
• question: "If you could only nail one thing, what would it be?"
• options: Key aspects they've mentioned + "All equally important" + "Something else"

4. Find boundaries:

Use AskUserQuestion:

• header: "Scope"
• question: "What's explicitly NOT in v1?"
• options: Things that might be tempting + "Nothing specific" + "Let me list them"

5. Ground in reality:

Use AskUserQuestion:

• header: "Constraints"
• question: "Any hard constraints?"
• options: Relevant constraint types + "None" + "Yes, let me explain"

6. Decision gate:

Use AskUserQuestion:

• header: "Ready?"
• question: "Ready to create PROJECT.md, or explore more?"
• options (ALL THREE REQUIRED):
  • "Create PROJECT.md" - Finalize and continue
  • "Ask more questions" - I'll dig deeper
  • "Let me add context" - You have more to share

If "Ask more questions" -> return to step 2. If "Let me add context" -> receive input via their response -> return to step 2. Loop until "Create PROJECT.md" selected.

</step> <step name="project">

Synthesize all context into .claude/dog/PROJECT.md.

PROJECT.md Structure:

# [Project Name]

## Overview
[One paragraph describing what this project is]

## Goals
- [Primary goal]
- [Secondary goals]

## Requirements

### Validated
(None yet - ship to validate)

### Active
- [ ] [Requirement 1]
- [ ] [Requirement 2]
- [ ] [Requirement 3]

### Out of Scope
- [Exclusion 1] - [why]
- [Exclusion 2] - [why]

## Constraints
- [Constraint 1]
- [Constraint 2]

## Key Decisions

| Decision | Rationale | Outcome |
|----------|-----------|---------|
| [Choice from questioning] | [Why] | Pending |

---
*Last updated: [date] after initialization*

Do not compress. Capture everything gathered.

</step> <step name="roadmap">

Create initial .claude/dog/ROADMAP.md:

# Roadmap

## Major 1: [First Major Version Name]
- **1.0:** [Initial minor version description]

---
*Use /dog:add-major-version to add more major versions*
*Use /dog:add-minor-version to add minor versions*
*Use /dog:add-task to add tasks to minor versions*

</step> <step name="mode">

Ask workflow mode preference:

Use AskUserQuestion:

• header: "Mode"
• question: "How do you want to work?"
• options:
  • "Interactive" - Confirm at each step, approval gates enabled
  • "YOLO" - Auto-approve, just execute

</step> <step name="config">

Create .claude/dog/dog-config.json:

{
  "yoloMode": [true if YOLO selected, false otherwise],
  "contextLimit": 200000,
  "targetContextUsage": 0.4,
  "autoCleanupWorktrees": true
}

</step> <step name="commit">

git add .claude/dog/PROJECT.md .claude/dog/ROADMAP.md .claude/dog/dog-config.json
git commit -m "$(cat <<'EOF'
docs: initialize [project-name]

[One-liner from PROJECT.md]

Creates PROJECT.md with requirements, ROADMAP.md, and config.
EOF
)"

</step> <step name="done">

Present completion with next steps:

Project initialized:

- Project: .claude/dog/PROJECT.md
- Roadmap: .claude/dog/ROADMAP.md
- Config: .claude/dog/dog-config.json (mode: [chosen mode])

---

## Next Up

**[Project Name]** - add first major version structure

`/dog:add-major-version`

<sub>`/clear` first -> fresh context window</sub>

---

</step> </process> <output>

• .claude/dog/PROJECT.md
• .claude/dog/ROADMAP.md
• .claude/dog/dog-config.json

</output>

<success_criteria>

• Deep questioning completed (not rushed)
• PROJECT.md captures full context with evolutionary structure
• Requirements initialized as hypotheses
• Key Decisions table initialized
• ROADMAP.md created with placeholder
• dog-config.json has workflow mode
• All committed to git

</success_criteria>