# Pattern Index - Continual Learning

> Read pattern → Follow checklist → Complete removal from dual systems (MCP + Extensions)

- Skill: `tools-only/pattern-index-continual-learning` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add tools-only/pattern-index-continual-learning`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tools-only/pattern-index-continual-learning/raw
- Safety review: pending (external: skill-scanner FAIL, skillspector CAUTION)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: tools-only (https://skillmd.com/u/tools-only)
- Updated: 2026-09-29
- Page: https://skillmd.com/skills/tools-only/pattern-index-continual-learning

---

# Pattern Index - Continual Learning

Quick reference for captured patterns and learning sessions.

**Last Updated:** 2025-11-01
**Total Patterns:** 2
**Total Sessions:** 1

---

## 🎯 Patterns (Reusable)

### 1. Claude Desktop Server Troubleshooting
**File:** `patterns/claude-desktop-troubleshooting.md`
**ID:** `claude-desktop-server-debug-001`
**Category:** System Debugging
**Token Savings:** 400-650 per use

**Use When:**
- Claude Desktop crashes
- "Server connection failed" errors
- Slow startup
- Unknown/problematic MCP servers or extensions

**Quick Action:**
Read pattern → Follow checklist → Complete removal from dual systems (MCP + Extensions)

---

### 2. Tool/Extension Evaluation Framework
**File:** `patterns/tool-evaluation-framework.md`
**ID:** `tool-evaluation-framework-001`
**Category:** Decision Making
**Token Savings:** 300-500 per use

**Use When:**
- Deciding to install new tool/server/extension
- Optimizing installed components
- Choosing between alternatives
- Performance/overhead concerns

**Quick Action:**
Read framework → Score 5 criteria → Apply decision matrix → Keep/Remove/Evaluate

---

## 📚 Learning Sessions (Historical)

### Session 1: Claude Desktop Troubleshooting & Configuration Optimization
**Date:** 2025-11-01
**File:** `sessions/2025-11-01-claude-desktop-troubleshooting.md`
**Duration:** ~45 minutes

**Key Outcomes:**
- ✅ Debugged AWS MCP Server crashes
- ✅ Discovered MCP vs Extensions dual system
- ✅ Created systematic troubleshooting checklist
- ✅ Developed tool evaluation framework
- ✅ Optimized configuration (9 → 4 servers)
- ✅ Installed 4 high-value Claude Code skills

**Patterns Extracted:**
1. Claude Desktop Server Troubleshooting
2. Tool/Extension Evaluation Framework

**ROI:**
- Time invested: 60 minutes
- Expected savings: 2.5-7.5 hours over next year
- Token savings: 2500-8000 tokens

---

## 🔍 Pattern Quick Reference

### By Category

**System Debugging:**
- `claude-desktop-server-debug-001` - Server troubleshooting

**Decision Making:**
- `tool-evaluation-framework-001` - Tool selection criteria

**[Future Categories]:**
- Code Architecture
- Deployment & DevOps
- Security & Compliance
- Documentation
- Testing & QA

---

## 📊 Usage Statistics

| Pattern ID | Uses | Avg Token Savings | Total Saved |
|------------|------|-------------------|-------------|
| claude-desktop-server-debug-001 | 0 | 500 | 0 |
| tool-evaluation-framework-001 | 0 | 400 | 0 |

**Note:** Update this table when patterns are reused!

---

## 🚀 How to Use This Index

### In a New Session:

1. **Problem Recognition:**
   - "Claude Desktop is crashing" → Check System Debugging category
   - "Should I install this tool?" → Check Decision Making category

2. **Pattern Lookup:**
   ```
   Find pattern in category → Open pattern file → Follow steps
   ```

3. **After Using Pattern:**
   - Update usage statistics
   - Note any improvements/modifications
   - Consider creating sub-patterns if needed

---

## 💡 Creating New Patterns

### When to Capture:
- Solved a complex problem (20+ minutes effort)
- Same approach used 2+ times
- Novel solution that generalizes
- Will save 300+ tokens on reuse

### How to Capture:
1. Create session file: `sessions/YYYY-MM-DD-[topic].md`
2. Extract patterns: `patterns/[pattern-name].md`
3. Update this index
4. Assign pattern ID: `[category]-[number]`

---

## 🔗 Related Resources

**Claude Code Skills:**
- `token-budget-advisor/` - Context optimization
- `security-analyzer/` - Security scanning
- `doc-generator/` - Documentation generation

**Configuration Files:**
- `~/.claude/settings.json`
- `~/.config/claude/claude_desktop_config.json`
- `~/projects/AGENT_MEMORY.md` (user's custom memory file)

---

## ✅ Next Steps

**High-Priority Patterns to Develop:**
1. React/Vite project setup best practices
2. API security implementation patterns
3. Data pipeline optimization
4. Git workflow automation
5. Production deployment checklist

**Trigger:** When working on these topics, use learning-capture skill to document!

---

**Maintained by:** Continual Learning Skill
**Updates:** Add entry when new pattern/session is created