Creates slash commands for Claude Code following Sando's Golden Rule - only create commands that add intelligent value over bash (analysis, insights, multi-source aggregation, troubleshooting). Use when user requests new slash command or workflow automation that justifies token cost.
Creates slash commands for Claude Code following Sando's Golden Rule - only create commands that add intelligent value over bash (analysis, insights, multi-source aggregation, troubleshooting). Use when user requests new slash command or workflow automation that justifies token cost.
allowed-tools
Read, Write, Edit, Glob, Bash
Command Creator Skill
This skill generates slash commands that follow Sando Design System philosophy: only create commands that justify token cost by adding intelligent value.
Guidelines: Single Source of Truth
CRITICAL: Command functionality must respect Sando guidelines when analyzing/validating project artifacts.
---
description: Brief description (shown in /help)
allowed-tools: Tool1, Tool2, Tool3
argument-hint: [arg-name (optional)] # If accepts arguments
---
Brief introduction.
# Section 1: Data Gathering
Explain what information to collect from where.
Reference guidelines for validation thresholds.
# Section 2: Analysis
Explain how to interpret data.
Apply guideline standards (TEST_COVERAGE.md thresholds, etc.).
# Section 3: Recommendations
Provide actionable recommendations.
# Output Format
Show example output format.
Command Creation Workflow
Step 1: Validate Against Golden Rule
# Check package.json scriptcat package.json | grep -A 50 '"scripts"' | grep "command-name"# If exists → STOP, use bash# If not exists → Continue
Skill shortcut? (explicit invocation of existing skill)
Performance analysis? (metrics + regressions)
If none → ❌ Command NOT justified
Step 3: Determine Command Location
Status/Info:.claude/commands/status/
Review/Analysis:.claude/commands/review/
Creation/Generation:.claude/commands/new/
Troubleshooting:.claude/commands/debug/
Performance:.claude/commands/performance/
Step 4: Create Command File
Using template structure with guideline references.
Step 5: Add Justification Comment
At end of command file:
---
## 💰 Token Cost Justification**Why this command is worth token cost:**- ✅ Combines X sources of information
- ✅ Analyzes per GUIDELINE.md standards (lines X-Y)
- ✅ Generates Z actionable recommendations
- ✅ Cannot be done with single bash command
**Estimated tokens per use:** ~XXX tokens
**Value added:** [Specific value]
**ROI:** Positive after X uses
"El mejor comando es el que no necesitas crear porque bash ya lo hace gratis."
Only create commands that justify token cost through intelligent analysis, multi-source aggregation, troubleshooting automation, or convenience shortcuts to existing skills.
When in doubt: "¿Puedo hacer esto con bash directamente?"
