Dex Backlog

# 📊 Backlog Review Complete

*Analyzed {{total_ideas}} ideas against current system state*

## 🔥 Top 5 Recommendations

### 1. [idea-XXX] {{title}} (Score: {{score}})

**Why now:** {{reasoning based on scores - be specific}}

**Quick assessment:**
- Impact: {{impact_justification}}
- Fits your patterns: {{alignment_justification}}
- Effort: {{effort_estimate}}

**Next step:** Run `/dex-improve "{{title}}"` to workshop this idea

---

### 2. [idea-YYY] {{title}} (Score: {{score}})

[Same format]

---

[... continue for top 5 ...]

---

## 📈 Backlog Health

- **Total ideas:** {{total}}
- **High priority (85+):** {{high_count}}
- **Medium priority (60-84):** {{medium_count}}
- **Low priority (<60):** {{low_count}}

{{#if high_count > 5}}
⚠️ **Note:** You have {{high_count}} high-priority ideas. Consider tackling 1-2 this week to reduce backlog.
{{/if}}

{{#if low_count > 10}}
💡 **Tip:** {{low_count}} low-priority ideas might be worth archiving or refining.
{{/if}}

---

## What would you like to do?

1. **Workshop an idea** → `/dex-improve "[title]"`
2. **Capture a new idea** → Use `capture_idea` MCP tool
3. **Mark one implemented** → Use `mark_implemented` MCP tool
4. **View full backlog** → Check `System/Dex_Backlog.md`

# 📊 Backlog Review

Your backlog is empty! 

Start capturing improvement ideas:
- Use the `capture_idea` MCP tool anytime you think "I wish Dex did X"
- Run `/dex-improve` to explore capability gaps
- Run `/dex-level-up` to discover unused features

The backlog system will help you track and prioritize ideas systematically.

🎉 **Good news:** No urgent improvements needed!

Your system is working well. The backlog has ideas for later, but nothing critical right now.

Consider:
- Running `/dex-level-up` to discover unused features
- Capturing ideas as they come up
- Reviewing backlog quarterly

⚠️ **Backlog maintenance needed**

You have {{stale_count}} ideas older than 6 months. These might be:
- No longer relevant → Archive them
- Still valuable but not urgent → Keep them
- Worth revisiting with new context → Re-evaluate descriptions

Review stale ideas:
{{list stale ideas}}

Want to bulk archive these? I can help clean up the backlog.

def check_cursor_feasibility(idea: dict) -> dict:
    """
    Returns: {
        'feasible': bool,
        'reason': str,
        'capabilities_required': list
    }
    """
    description_lower = idea['description'].lower()
    
    # Red flags - things Cursor CAN'T do
    cannot_do = {
        'track edits': 'Cannot monitor file edits in real-time',
        'watch user': 'Cannot observe user actions passively',
        'hook into': 'Cannot hook into Cursor internals',
        'monitor changes': 'Cannot monitor without explicit file reads',
        'background process': 'No persistent background processes'
    }
    
    for phrase, reason in cannot_do.items():
        if phrase in description_lower:
            return {
                'feasible': False,
                'reason': reason,
                'suggestion': 'Reframe as file-based or command-triggered'
            }
    
    # Green flags - things Cursor CAN do
    can_do = ['file', 'read', 'write', 'mcp', 'command', 'cache', 'index', 'parse']
    has_feasible_approach = any(word in description_lower for word in can_do)
    
    if has_feasible_approach:
        return {'feasible': True, 'reason': 'Uses Cursor-compatible operations'}
    else:
        return {
            'feasible': False,
            'reason': 'No clear implementation path in Cursor',
            'suggestion': 'Add file-based or MCP approach'
        }

def calculate_impact(idea, context):
    # First check feasibility
    feasibility = check_cursor_feasibility(idea)
    if not feasibility['feasible']:
        return 0  # Not feasible = 0 impact
    
    score = 50
    
    # Check session learnings for pain point mentions
    learnings = context['session_learnings']
    idea_keywords = extract_keywords(idea['title'] + idea['description'])
    
    for learning in learnings:
        learning_keywords = extract_keywords(learning['content'])
        if overlap(idea_keywords, learning_keywords) > 0.3:
            score += 20
            break
    
    # Check if affects daily workflow
    if touches_daily_commands(idea, context['usage_log']):
        score += 15
    
    # Check compound value
    if enables_other_ideas(idea, context['backlog']):
        score += 15
    
    return min(score, 100)

def calculate_alignment(idea, context):
    score = 50
    
    # Extract related features
    features = extract_related_features(idea)
    used_features = get_used_features(context['usage_log'])
    
    overlap_ratio = len(features & used_features) / len(features)
    score += int(overlap_ratio * 30)
    
    # Role fit
    role = context['user_profile']['role']
    category = idea['category']
    
    role_fit_map = {
        'PM': ['projects', 'workflows', 'knowledge'],
        'Sales': ['relationships', 'tasks'],
        'Leadership': ['knowledge', 'workflows']
    }
    
    if category in role_fit_map.get(role, []):
        score += 20
    
    return min(score, 100)

def calculate_token_efficiency(idea, context):
    score = 50
    
    # Check if reduces token usage
    if reduces_reads(idea):  # Caching, summaries
        score += 25
    
    # Context efficiency improvements
    if improves_retrieval(idea):  # Better indexing, structured data
        score += 15
    
    # Incremental updates
    if supports_incremental(idea):  # Partial updates, lazy loading
        score += 10
    
    return min(score, 100)

def calculate_memory_learning(idea, context):
    score = 50
    
    # Memory persistence
    if stores_learnings(idea):  # Knowledge base, historical context
        score += 20
    
    # Self-learning capability
    if enables_adaptation(idea):  # Learns from behavior, improves over time
        score += 20
    
    # Feedback loops
    if tracks_outcomes(idea):  # Measures effectiveness, refines
        score += 10
    
    return min(score, 100)

def calculate_proactivity(idea, context):
    score = 50
    
    # Anticipation capability
    if enables_anticipation(idea):  # Surfaces info before asked
        score += 25
    
    # Automation of routine decisions
    if automates_decisions(idea):  # Handles repetitive choices
        score += 15
    
    # Timing improvements
    if improves_timing(idea):  # Right info at right time
        score += 10
    
    return min(score, 100)