This skill should be used when the user asks to "plan a feature", "evaluate a feature", "review a feature proposal", "analyze a feature before implementation", "plan a feature", "review this feature", "assess this feature", or uses the /super-dev command for multi-agent feature planning with 5-perspective judgment system.
View on GitHubsuper-dev/skills/super-dev/SKILL.md
February 1, 2026
Select agents to install to:
npx add-skill https://github.com/JuniYadi/claude-code/blob/main/super-dev/skills/super-dev/SKILL.md -a claude-code --skill super-devInstallation paths:
.claude/skills/super-dev/# Super-Dev: Multi-Agent Feature Planning
Multi-agent feature planning with 5-perspective judgement system for validating new features before implementation.
## ⚠️ CRITICAL: Role Definition
**This skill is PLANNING and INSPECTION ONLY:**
- ✅ Analyze proposals
- ✅ Evaluate with 5 judges
- ✅ Provide recommendations
- ❌ Do NOT implement features
- ❌ Do NOT make code changes
- ❌ Do NOT create files
**Output Requirement:**
When running as a sub-agent, you MUST output the complete formatted report with ALL judge verdicts. Do not summarize. Do not internalize. OUTPUT the full report so the user can see it.
**Stop After Planning:**
After presenting the final report, STOP. Do not ask if the user wants to implement. Do not offer to implement. Your job is done after evaluation.
## Purpose
Transform vague feature requests into structured proposals, then evaluate them through 5 independent judge agents (Pragmatist, Purist, Innovator, Skeptic, Optimizer) to produce actionable go/no-go decisions.
## Usage
```bash
# With vague request (triggers interactive proposal builder)
/super-dev add notifications to Laravel app
# With file (structured proposal)
/super-dev --file docs/proposals/notification-system.md
```
## High-Level Flow
```
User Request
↓
Phase 0: Proposal Building (if needed)
↓
Phase 1: Parse & Analyze
↓
Phase 2: 5 Judges in PARALLEL
↓
Phase 3: Vote Aggregation
↓
Phase 4: Present Final Report
```
## Implementation
### Phase 0: Input Analysis & Proposal Building
```markdown
Analyze user input to determine if proposal builder is needed.
**Check for completeness:**
- Does input have structured sections (Goals, Requirements, Constraints)?
- Is it just a vague request like "add notifications"?
- Is it a file path with --file flag?
**Decision flow:**
if args starts with "--file":
proposal_file = args after "--file"
if file exists:
proposal_text = read file
→ Skip to Phase 1
else:
ERROR: "File not found: {p