ralph

npx skills add majiayu000/claude-skill-registry --skill ralph-doravidan-supreme-ralph-3 --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

# Ralph Skill Convert PRD Markdown files to prd.json format for RALPH automation. ## Trigger This skill activates when the user asks to: - Convert a PRD to JSON - Prepare a PRD for RALPH - Create prd.json from a PRD file ## Process ### Step 1: Read the PRD Read the specified PRD Markdown file: ```bash cat tasks/prd-[feature-name].md ``` ### Step 2: Extract Information Parse the PRD to extract: - Project name (from title) - Description (from overview) - User stories with: - ID (US-001, US-002, etc.) - Title - Description (As a... I want... So that...) - Acceptance criteria - Priority ### Step 3: Generate prd.json Create the JSON structure: ```json { "project": "[Feature Name]", "branchName": "ralph/[feature-slug]", "description": "[Overview text]", "createdAt": "[Today's date]", "userStories": [ { "id": "US-001", "title": "[Story title]", "description": "[Full story description]", "acceptanceCriteria": [ "[Criterion 1]", "[Criterion 2]", "Typecheck passes" ], "priority": 1, "passes": false, "notes": "" } ] } ``` ### Step 4: Validate and Save Before saving, validate: 1. Every story has "Typecheck passes" in criteria 2. Stories are ordered by priority (dependencies first) 3. Each stor

cat tasks/prd-[feature-name].md
if [ -f prd.json ]; then
mkdir -p archive/$(date +%Y-%m-%d)
cp prd.json archive/$(date +%Y-%m-%d)/
cp progress.txt archive/$(date +%Y-%m-%d)/ 2>/dev/null || true
fi
cat > prd.json << 'EOF'
cat > progress.txt << 'EOF'