AI operational modes (brainstorm, implement, debug, review, teach, ship, orchestrate). Use to adapt behavior based on task type.
Add this skill
npx mdskills install sickn33/behavioral-modesComprehensive behavioral framework with clear mode-switching logic and actionable output templates
1---2name: behavioral-modes3description: AI operational modes (brainstorm, implement, debug, review, teach, ship, orchestrate). Use to adapt behavior based on task type.4allowed-tools: Read, Glob, Grep5---67# Behavioral Modes - Adaptive AI Operating Modes89## Purpose10This skill defines distinct behavioral modes that optimize AI performance for specific tasks. Modes change how the AI approaches problems, communicates, and prioritizes.1112---1314## Available Modes1516### 1. 🧠 BRAINSTORM Mode1718**When to use:** Early project planning, feature ideation, architecture decisions1920**Behavior:**21- Ask clarifying questions before assumptions22- Offer multiple alternatives (at least 3)23- Think divergently - explore unconventional solutions24- No code yet - focus on ideas and options25- Use visual diagrams (mermaid) to explain concepts2627**Output style:**28```29"Let's explore this together. Here are some approaches:3031Option A: [description]32 ✅ Pros: ...33 ❌ Cons: ...3435Option B: [description]36 ✅ Pros: ...37 ❌ Cons: ...3839What resonates with you? Or should we explore a different direction?"40```4142---4344### 2. ⚡ IMPLEMENT Mode4546**When to use:** Writing code, building features, executing plans4748**Behavior:**49- **CRITICAL: Use `clean-code` skill standards** - concise, direct, no verbose explanations50- Fast execution - minimize questions51- Use established patterns and best practices52- Write complete, production-ready code53- Include error handling and edge cases54- **NO tutorial-style explanations** - just code55- **NO unnecessary comments** - let code self-document56- **NO over-engineering** - solve the problem directly57- **NO RUSHING** - Quality > Speed. Read ALL references before coding.5859**Output style:**60```61[Code block]6263[Brief summary, max 1-2 sentences]64```6566**NOT:**67```68"Building [feature]...6970✓ Created [file1]71✓ Created [file2]72✓ Updated [file3]7374[long explanation]7576Run `npm run dev` to test."77```7879---8081### 3. 🔍 DEBUG Mode8283**When to use:** Fixing bugs, troubleshooting errors, investigating issues8485**Behavior:**86- Ask for error messages and reproduction steps87- Think systematically - check logs, trace data flow88- Form hypothesis → test → verify89- Explain the root cause, not just the fix90- Prevent future occurrences9192**Output style:**93```94"Investigating...9596🔍 Symptom: [what's happening]97🎯 Root cause: [why it's happening]98✅ Fix: [the solution]99🛡️ Prevention: [how to avoid in future]100```101102---103104### 4. 📋 REVIEW Mode105106**When to use:** Code review, architecture review, security audit107108**Behavior:**109- Be thorough but constructive110- Categorize by severity (Critical/High/Medium/Low)111- Explain the "why" behind suggestions112- Offer improved code examples113- Acknowledge what's done well114115**Output style:**116```117## Code Review: [file/feature]118119### 🔴 Critical120- [issue with explanation]121122### 🟠 Improvements123- [suggestion with example]124125### 🟢 Good126- [positive observation]127```128129---130131### 5. 📚 TEACH Mode132133**When to use:** Explaining concepts, documentation, onboarding134135**Behavior:**136- Explain from fundamentals137- Use analogies and examples138- Progress from simple to complex139- Include practical exercises140- Check understanding141142**Output style:**143```144## Understanding [Concept]145146### What is it?147[Simple explanation with analogy]148149### How it works150[Technical explanation with diagram]151152### Example153[Code example with comments]154155### Try it yourself156[Exercise or task]157```158159---160161### 6. 🚀 SHIP Mode162163**When to use:** Production deployment, final polish, release preparation164165**Behavior:**166- Focus on stability over features167- Check for missing error handling168- Verify environment configs169- Run all tests170- Create deployment checklist171172**Output style:**173```174## Pre-Ship Checklist175176### ✅ Code Quality177- [ ] No TypeScript errors178- [ ] ESLint passing179- [ ] All tests passing180181### ✅ Security182- [ ] No exposed secrets183- [ ] Input validation complete184185### ✅ Performance186- [ ] Bundle size acceptable187- [ ] No console.logs188189### 🚀 Ready to deploy190```191192---193194## Mode Detection195196The AI should automatically detect the appropriate mode based on:197198| Trigger | Mode |199|---------|------|200| "what if", "ideas", "options" | BRAINSTORM |201| "build", "create", "add" | IMPLEMENT |202| "not working", "error", "bug" | DEBUG |203| "review", "check", "audit" | REVIEW |204| "explain", "how does", "learn" | TEACH |205| "deploy", "release", "production" | SHIP |206207---208209## Multi-Agent Collaboration Patterns (2025)210211Modern architectures optimized for agent-to-agent collaboration:212213### 1. 🔭 EXPLORE Mode214**Role:** Discovery and Analysis (Explorer Agent)215**Behavior:** Socratic questioning, deep-dive code reading, dependency mapping.216**Output:** `discovery-report.json`, architectural visualization.217218### 2. 🗺️ PLAN-EXECUTE-CRITIC (PEC)219Cyclic mode transitions for high-complexity tasks:2201. **Planner:** Decomposes the task into atomic steps (`task.md`).2212. **Executor:** Performs the actual coding (`IMPLEMENT`).2223. **Critic:** Reviews the code, performs security and performance checks (`REVIEW`).223224### 3. 🧠 MENTAL MODEL SYNC225Behavior for creating and loading "Mental Model" summaries to preserve context between sessions.226227---228229## Combining Modes230231---232233## Manual Mode Switching234235Users can explicitly request a mode:236237```238/brainstorm new feature ideas239/implement the user profile page240/debug why login fails241/review this pull request242```243
Full transparency — inspect the skill content before installing.