superclaw

# Superclaw ⚔️

**Disciplined software development workflow for OpenClaw agents**

Based on [obra/superpowers](https://github.com/obra/superpowers) by Jesse Vincent.

---

## What This Skill Package Does

Superclaw prevents your agent from jumping straight into code. It enforces a three-phase workflow:

1. **🧠 Brainstorming** (`brainstorming/SKILL.md`) — Design before code
2. **📋 Writing Plans** (`writing-plans/SKILL.md`) — Plan before implementation  
3. **⚙️ Executing Plans** (`executing-plans/SKILL.md`) — Batched execution with checkpoints

All three skills chain automatically when building software.

---

## How It Works

### Phase 1: Brainstorming (Design Before Code)

**Triggers:** When creating features, building components, adding functionality

**Process:**
1. Check context (MEMORY.md, USER.md, daily logs)
2. Ask Socratic questions (requirements, constraints, trade-offs)
3. Propose 2-3 approaches with pros/cons
4. Present design
5. Get approval
6. Save design document to `workspace/docs/plans/YYYY-MM-DD-<topic>-design.md`
7. **Automatically invoke writing-plans skill**

**Hard Gate:** No code until design approved.

---

### Phase 2: Writing Plans (Plan Before Implementation)

**Triggers:** When you have an approved design

**Process:**
1. **ASK about methodology** (TDD? Direct implementation?)
2. Ask about commit frequency
3. Break work into 2-5 minute tasks
4. Save implementation plan to `workspace/docs/plans/YYYY-MM-DD-<topic>-plan.md`
5. **Automatically invoke executing-plans skill**

**Key Feature:** Questions, not mandates. Respects user preferences and time constraints.

---

### Phase 3: Executing Plans (Batched Execution with Checkpoints)

**Triggers:** When you have an implementation plan

**Process:**
1. Load plan from document
2. Batch tasks into groups of 3-5
3. Execute batch (using `sessions_spawn` for isolation)
4. Review outputs
5. Checkpoint ("Batch N complete. Continue?")
6. Update `memory/YYYY-MM-DD.md` with progress
7. Repeat until com...

# Superclaw ⚔️

**A complete software development workflow for OpenClaw agents**

Superclaw enforces discipline through three chained skills: design before code, plan before implementation, batched execution with checkpoints.

Based on [obra/superpowers](https://github.com/obra/superpowers) by Jesse Vincent, adapted for OpenClaw's personal assistant architecture.

---

## What It Does

Your agent doesn't just jump into code. It:

1. **🧠 Brainstorms with you** — Socratic questions, design proposals, approval before coding
2. **📋 Writes TDD-ready plans** — Breaks work into 2-5 minute tasks (with methodology questions, not mandates)
3. **⚙️ Executes in batches** — Groups tasks, checkpoints between batches, updates memory

---

## Installation (Future)

```bash
openclaw hub install superclaw
```

Skills auto-load when relevant tasks are detected.

---

## Skills Included

### 1. Brainstorming (`brainstorming/SKILL.md`)
**Enforces design-before-code**

- **Triggers:** When creating features, building components, adding functionality
- **Hard Gate:** No code until design approved
- **Process:** Context check → Questions → Approaches → Design → Approval → Save
- **Then:** Automatically invokes writing-plans skill

**Example:**
```
User: "Build a todo CLI"
Agent: [Asks questions about storage, features]
Agent: [Proposes 2-3 approaches with trade-offs]
Agent: [Presents design, gets approval]
Agent: [Saves design doc, invokes writing-plans]
```

### 2. Writing-Plans (`writing-plans/SKILL.md`)
**Enforces plan-before-implementation**

- **Triggers:** When you have an approved design
- **Key Feature:** ASKS about methodology (TDD? Commit frequency?) instead of mandating
- **Process:** Ask methodology → Break into tasks → Save plan
- **Then:** Automatically invokes executing-plans skill

**Example:**
```
Agent: "Should we use TDD (tests first) or direct implementation?"
User: "Direct implementation"
Agent: [Creates plan with 10 tasks, 2-5 min each]
Agent: [Saves plan doc, invok...