# AI Learned Rules (From User Feedback)

**Purpose**: Persistent memory of user corrections and preferences  
**Usage**: AI MUST read this file at start of EVERY workflow  
**Update**: Add new rules whenever user provides corrections  
**Priority**: CRITICAL rules cannot be violated

---

## 🔴 Critical Rules (NEVER Violate)

### RULE-001: [Your First Critical Rule]

**Learned From**: [Session/Date when you learned this]  
**User Quote**: "[Exact user feedback that triggered this rule]"

**The Rule**:
[Clear statement of what AI MUST or MUST NOT do]

**Why This Rule Exists**:
- [Problem it solves]
- [Pain point it addresses]
- [Impact if violated]

**Actions Required**:
1. ✅ [Specific action step 1]
2. ✅ [Specific action step 2]
3. ✅ [Specific action step 3]

**Examples**:

**Good Example** ✅:
```
[Show correct behavior]
```

**Bad Example** ❌:
```
[Show incorrect behavior to avoid]
```

**Enforcement**:
- [How this rule is validated]
- [What prevents violations]
- [Consequences of violation]

**Related**: [Links to other rules or documentation]

---

## 🟡 High Priority Rules

### RULE-002: [Your Second Rule]

[Follow same structure as above]

---

## 📚 How to Use This File

### **For AI Agents** (MANDATORY):

**At Start of EVERY Task**:
```
1. Read docs/learned-knowledge/AI_LEARNED_RULES.md (this file)
2. Read docs/learned-knowledge/USER_PREFERENCES.md
3. Identify relevant rules for current task
4. Show user: "Applying learned rules: X, Y, Z"
5. Execute with rules in mind
6. After completion: Check if new rules learned
7. Update files if user provided corrections
```

**During Execution**:
```
Before critical decisions:
  → Check: "Any learned rule about this?"
  → Apply: Rule if found
  → Proceed: With rule in mind
```

**After User Correction**:
```
1. Acknowledge correction
2. Extract rule from feedback
3. Add to this file
4. Create memory (long-term storage)
5. Show user: "Learned: [rule summary]"
```

---

## 📈 Rule Effectiveness Tracking

| Rule ID | Times Applied | Violations Prevented | Last Applied |
|---------|---------------|---------------------|--------------|
| RULE-001 | 0 | 0 | N/A |

**Update this table** when rules are applied or violated.

---

**This file is the AI's LONG-TERM MEMORY.** 🧠  
**Read it. Apply it. Update it. Learn from it.**