[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6D341GH59DBVQCC7QJQB5M9":3,"skill-same:01M6D341GH59DBVQCC7QJQB5M9":51},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":23,"linkOnly":24,"markdownContent":25,"markdownBody":26,"frontMatter":27,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":34,"ownerDescription":35,"updatedAt":36,"install":37,"resources":50},"01M6D341GH59DBVQCC7QJQB5M9","amp-backlog",null,"AI-powered ranking of Amp system improvement ideas","hemory\u002Famp-legacy-do-not-use","hemory","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F21179881?v=4","01M6D31HFJ391DN8N0JP8AQGFE","Hemory Phifer","01M6D31HKXVTEN27AT6R4PP2ZJ",3,"MIT","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Famp-backlog\u002FSKILL.md",[],[19,20,21,22],"backlog","amp backlog","system","ideas","2026-11-10 10:12:30",false,"---\nname: amp-backlog\ndescription: AI-powered ranking of Amp system improvement ideas\n---\n\n## What This Command Does\n\n**In plain English:** AI-powered ranking of your Amp system improvement backlog based on current system state. Shows you what to build next.\n\n**When to use it:**\n- Weekly check-in on system improvements\n- After capturing several new ideas\n- When deciding what to work on next\n- During quarterly planning for system improvements\n\n**How to run it:**\n```\n\u002Famp-backlog              # Full review with re-ranking\n```\n\n---\n\n## Process Overview\n\n1. **Load context** - Read system state, usage patterns, learnings\n2. **Score ideas** - Calculate 5-dimension scores for each idea\n3. **Re-rank backlog** - Sort by weighted total score\n4. **Update file** - Write new rankings to `System\u002FAmp_Backlog.md`\n5. **Present top ideas** - Show top 5 with \"Why now?\" justification\n6. **Offer next steps** - Workshop, implement, or defer\n\n---\n\n## Step 1: Load System Context\n\nRead these files to understand current system state:\n\n### Required Files\n```\nSystem\u002FAmp_Backlog.md              # All ideas to score\nSystem\u002Fusage_log.md                # Feature adoption patterns\nSystem\u002Fuser-profile.yaml           # Role, preferences\nCLAUDE.md                          # Current capabilities\n```\n\n### Optional Files (if they exist)\n```\nSystem\u002FSession_Learnings\u002F           # Recent pain points (last 30 days)\n.claude\u002Fcommands\u002F                  # Available commands\ncore\u002Fmcp\u002F                          # MCP integrations\n06-Resources\u002FLearnings\u002F               # Captured patterns\n```\n\n### Extract Context\n\nBuild a context dictionary with:\n- **Usage patterns**: Which features are used vs. unused\n- **Role profile**: PM, Sales, Leadership, Engineer, etc.\n- **Pain points**: Recent friction from session learnings\n- **System capabilities**: What's currently available\n- **Backlog state**: All ideas and their current scores\n\n---\n\n## Step 2: Score Each Idea\n\nFor every active idea in the backlog, calculate 5 dimension scores.\n\n### ⚠️ CURSOR FEASIBILITY CHECK (Do This First!)\n\nBefore scoring ANY idea, validate it's actually implementable in Cursor:\n\n**What Cursor\u002FTerminal CAN do:**\n- ✅ Read and write files\n- ✅ Execute shell commands\n- ✅ Build MCP tools for structured operations\n- ✅ Parse and transform file contents\n- ✅ Create caches and indexes (file-based)\n- ✅ Run commands on schedules or triggers\n\n**What Cursor\u002FTerminal CANNOT do:**\n- ❌ Track user edits in real-time\n- ❌ Hook into Cursor internals\n- ❌ Monitor user actions passively\n- ❌ Access edit history without explicit file reads\n- ❌ Real-time background processes watching for changes\n\n**If idea requires something from the CANNOT list → Set all scores to 0 and flag as \"Not feasible in Cursor\"**\n\n---\n\nAfter feasibility check passes, score on 5 dimensions:\n\n### Dimension 1: Impact (35% weight)\n\n**Question:** How much would this improve daily workflow?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+20 if matches_recent_pain_points():\n  - Search System\u002FSession_Learnings\u002F for mentions of this issue\n  - Keywords from idea title\u002Fdescription appear in learnings\n  - Problem stated explicitly in recent notes\n\n+15 if affects_daily_workflow():\n  - Touches commands used >3x per week (from usage_log)\n  - Modifies core files (03-Tasks\u002FTasks.md, daily plans, person pages)\n  - Impacts repetitive actions\n\n+15 if has_compound_value():\n  - Enables other ideas in backlog\n  - Reduces technical debt\n  - Creates reusable patterns\n  - Unblocks multiple workflows\n\nMax: 100\n```\n\n**Examples:**\n- \"Auto-suggest person pages\": 95 (daily workflow + enables relationship tracking)\n- \"Export to blog\": 40 (nice-to-have, doesn't affect core workflow)\n\n---\n\n### Dimension 2: Alignment (20% weight)\n\n**Question:** Does this fit actual usage patterns?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+30 based on usage_overlap():\n  - Extract features idea depends on\n  - Check if those features are used (usage_log)\n  - Calculate overlap: (used_features \u002F total_features) * 30\n\n+20 if fits_role_profile():\n  - PM roles: prioritize project\u002Fproduct features\n  - Sales roles: prioritize relationship\u002Faccount features\n  - Leadership: prioritize synthesis\u002Freview features\n  - Match category to role focus areas\n\nMax: 100\n```\n\n**Examples:**\n- Idea needs \"person pages\" → Check if user has created person pages\n- If usage_log shows person pages = used → Higher alignment\n- If role = PM and idea = product features → +20 role fit\n\n---\n\n### Dimension 3: Token Efficiency (20% weight)\n\n**Question:** Does this reduce context\u002Ftoken usage?\n\n**CRITICAL - Cursor Feasibility Check:**\nBefore scoring, verify the idea is implementable in Cursor\u002FTerminal:\n- ✅ Can use: File read\u002Fwrite, MCP tools, command execution, file-based caching\n- ❌ Cannot use: Real-time edit tracking, Cursor internal hooks, monitoring user actions\n- If not feasible in Cursor → Score = 0 on all dimensions\n\n**Scoring logic:**\n```\nBase score: 50\n\n+25 if reduces_token_usage():\n  - Caches\u002Fstores frequently accessed data (in files\u002FMCP)\n  - Compresses or summarizes verbose content (file-based)\n  - Eliminates redundant reads\n  - Enables more efficient retrieval patterns\n\n+15 if improves_context_efficiency():\n  - Reduces number of files that need reading\n  - Creates structured summaries (YAML\u002FJSON files)\n  - Better indexing\u002Fsearch to avoid broad scans\n  - Moves data from markdown to structured format\n\n+10 if enables_incremental_updates():\n  - Supports partial updates instead of full rewrites\n  - Tracks changes in separate files\n  - Lazy loading or on-demand computation\n\nMax: 100\n```\n\n**Examples:**\n- \"Cache meeting summaries in YAML\": 90 (file-based, avoids re-reading)\n- \"Track user edits for learning\": 0 (NOT FEASIBLE - can't track edits)\n- \"Add new field to template\": 50 (neutral token impact)\n\n---\n\n### Dimension 4: Memory & Learning (15% weight)\n\n**Question:** Does this enhance system memory, persistence, or self-learning?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+20 if improves_memory_persistence():\n  - Stores learnings for future reference\n  - Creates retrievable knowledge base\n  - Captures patterns that compound over time\n  - Builds historical context\n\n+20 if enables_self_learning():\n  - System learns from user behavior\n  - Adapts recommendations based on patterns\n  - Builds preference models\n  - Improves predictions over time\n\n+10 if creates_feedback_loops():\n  - Tracks outcomes of suggestions\n  - Measures effectiveness of recommendations\n  - Refines based on what works\n\nMax: 100\n```\n\n**Examples:**\n- \"Learning pattern synthesizer\": 95 (captures + compounds knowledge)\n- \"Preference learning from edits\": 85 (system adapts over time)\n- \"Static template update\": 50 (no learning component)\n\n---\n\n### Dimension 5: Proactivity (10% weight)\n\n**Question:** Does this enable proactive concierge behavior?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+25 if enables_anticipation():\n  - Surfaces relevant info before asked\n  - Predicts needs based on patterns\n  - Proactive suggestions not just reactive\n  - Context-aware prompts\n\n+15 if automates_routine_decisions():\n  - Handles repetitive choices automatically\n  - Learns user preferences and applies them\n  - Reduces decision fatigue\n\n+10 if improves_timing():\n  - Right information at right time\n  - Context-aware interruptions\n  - Anticipates upcoming needs\n\nMax: 100\n```\n\n**Examples:**\n- \"Auto-prep meetings based on calendar\": 90 (proactive + anticipatory)\n- \"Suggest weekly priorities from patterns\": 80 (learns and anticipates)\n- \"Add manual review step\": 50 (reactive, not proactive)\n\n---\n\n## Step 3: Calculate Weighted Score\n\n```\ntotal_score = (\n  (impact * 0.35) +\n  (alignment * 0.20) +\n  (token_efficiency * 0.20) +\n  (memory_learning * 0.15) +\n  (proactivity * 0.10)\n)\n\nRound to integer: total_score = round(total_score)\n```\n\n**Priority Bands:**\n- **High Priority (85+):** Should tackle soon, high ROI\n- **Medium Priority (60-84):** Good ideas, right time matters\n- **Low Priority (\u003C60):** Maybe later or needs refinement\n\n**Why These Dimensions:**\n- **Effort excluded:** With AI coding, implementation is cheap - focus on value, not cost\n- **Token efficiency prioritized:** Context efficiency is critical for performance\n- **Memory & learning emphasized:** System should get smarter over time\n- **Proactivity valued:** Concierge behavior > reactive tool\n\n---\n\n## Step 4: Update Backlog File\n\nRewrite `System\u002FAmp_Backlog.md` with:\n\n1. **Update timestamp** at top\n2. **Re-sort ideas** by total score (high to low)\n3. **Update each idea** with new scores:\n   ```markdown\n   - **[idea-XXX]** Title\n     - **Score:** 92 (Impact: 95, Alignment: 90, Effort: 85, Synergy: 95, Fresh: 70)\n     - **Category:** category\n     - **Captured:** YYYY-MM-DD\n     - **Why ranked here:** [1-2 sentence reasoning based on scores]\n     - **Description:** [original description]\n   ```\n4. **Place in correct section** (High\u002FMedium\u002FLow priority)\n5. **Preserve Archive section** (don't re-rank implemented ideas)\n\n---\n\n## Step 5: Present Results\n\nShow the user the top 5 ideas with context:\n\n```markdown\n# ? Backlog Review Complete\n\n*Analyzed {{total_ideas}} ideas against current system state*\n\n## ? Top 5 Recommendations\n\n### 1. [idea-XXX] {{title}} (Score: {{score}})\n\n**Why now:** {{reasoning based on scores - be specific}}\n\n**Quick assessment:**\n- Impact: {{impact_justification}}\n- Fits your patterns: {{alignment_justification}}\n- Effort: {{effort_estimate}}\n\n**Next step:** Run `\u002Famp-improve \"{{title}}\"` to workshop this idea\n\n---\n\n### 2. [idea-YYY] {{title}} (Score: {{score}})\n\n[Same format]\n\n---\n\n[... continue for top 5 ...]\n\n---\n\n## ? Backlog Health\n\n- **Total ideas:** {{total}}\n- **High priority (85+):** {{high_count}}\n- **Medium priority (60-84):** {{medium_count}}\n- **Low priority (\u003C60):** {{low_count}}\n\n{{#if high_count > 5}}\n⚠️ **Note:** You have {{high_count}} high-priority ideas. Consider tackling 1-2 this week to reduce backlog.\n{{\u002Fif}}\n\n{{#if low_count > 10}}\n? **Tip:** {{low_count}} low-priority ideas might be worth archiving or refining.\n{{\u002Fif}}\n\n---\n\n## What would you like to do?\n\n1. **Workshop an idea** → `\u002Famp-improve \"[title]\"`\n2. **Capture a new idea** → Use `capture_idea` MCP tool\n3. **Mark one implemented** → Use `mark_implemented` MCP tool\n4. **View full backlog** → Check `System\u002FAmp_Backlog.md`\n```\n\n---\n\n## Step 6: Handle Special Cases\n\n### If Backlog is Empty\n```markdown\n# ? Backlog Review\n\nYour backlog is empty! \n\nStart capturing improvement ideas:\n- Use the `capture_idea` MCP tool anytime you think \"I wish Amp did X\"\n- Run `\u002Famp-improve` to explore capability gaps\n- Run `\u002Famp-level-up` to discover unused features\n\nThe backlog system will help you track and prioritize ideas systematically.\n```\n\n### If No High Priority Ideas\n```markdown\n? **Good news:** No urgent improvements needed!\n\nYour system is working well. The backlog has ideas for later, but nothing critical right now.\n\nConsider:\n- Running `\u002Famp-level-up` to discover unused features\n- Capturing ideas as they come up\n- Reviewing backlog quarterly\n```\n\n### If Many Stale Ideas (>6 months old)\n```markdown\n⚠️ **Backlog maintenance needed**\n\nYou have {{stale_count}} ideas older than 6 months. These might be:\n- No longer relevant → Archive them\n- Still valuable but not urgent → Keep them\n- Worth revisiting with new context → Re-evaluate descriptions\n\nReview stale ideas:\n{{list stale ideas}}\n\nWant to bulk archive these? I can help clean up the backlog.\n```\n\n---\n\n## Integration with Other Commands\n\n### Hand-off to \u002Famp-improve\n\nWhen user says \"Let's work on #1\" or \"Workshop idea-XXX\":\n\n1. Read the idea details from backlog\n2. Pass to `\u002Famp-improve` with context:\n   ```\n   \u002Famp-improve \"{{idea_title}}\"\n   \n   Context from backlog:\n   - Current score: {{score}}\n   - Why it's prioritized: {{reasoning}}\n   - Original description: {{description}}\n   ```\n3. `\u002Famp-improve` takes over for workshopping\n\n---\n\n## Scoring Implementation Tips\n\n### Cursor Feasibility Check (Run FIRST)\n\n```python\ndef check_cursor_feasibility(idea: dict) -> dict:\n    \"\"\"\n    Returns: {\n        'feasible': bool,\n        'reason': str,\n        'capabilities_required': list\n    }\n    \"\"\"\n    description_lower = idea['description'].lower()\n    \n    # Red flags - things Cursor CAN'T do\n    cannot_do = {\n        'track edits': 'Cannot monitor file edits in real-time',\n        'watch user': 'Cannot observe user actions passively',\n        'hook into': 'Cannot hook into Cursor internals',\n        'monitor changes': 'Cannot monitor without explicit file reads',\n        'background process': 'No persistent background processes'\n    }\n    \n    for phrase, reason in cannot_do.items():\n        if phrase in description_lower:\n            return {\n                'feasible': False,\n                'reason': reason,\n                'suggestion': 'Reframe as file-based or command-triggered'\n            }\n    \n    # Green flags - things Cursor CAN do\n    can_do = ['file', 'read', 'write', 'mcp', 'command', 'cache', 'index', 'parse']\n    has_feasible_approach = any(word in description_lower for word in can_do)\n    \n    if has_feasible_approach:\n        return {'feasible': True, 'reason': 'Uses Cursor-compatible operations'}\n    else:\n        return {\n            'feasible': False,\n            'reason': 'No clear implementation path in Cursor',\n            'suggestion': 'Add file-based or MCP approach'\n        }\n```\n\n### For Impact Calculation\n\n```python\ndef calculate_impact(idea, context):\n    # First check feasibility\n    feasibility = check_cursor_feasibility(idea)\n    if not feasibility['feasible']:\n        return 0  # Not feasible = 0 impact\n    \n    score = 50\n    \n    # Check session learnings for pain point mentions\n    learnings = context['session_learnings']\n    idea_keywords = extract_keywords(idea['title'] + idea['description'])\n    \n    for learning in learnings:\n        learning_keywords = extract_keywords(learning['content'])\n        if overlap(idea_keywords, learning_keywords) > 0.3:\n            score += 20\n            break\n    \n    # Check if affects daily workflow\n    if touches_daily_commands(idea, context['usage_log']):\n        score += 15\n    \n    # Check compound value\n    if enables_other_ideas(idea, context['backlog']):\n        score += 15\n    \n    return min(score, 100)\n```\n\n### For Alignment Calculation\n\n```python\ndef calculate_alignment(idea, context):\n    score = 50\n    \n    # Extract related features\n    features = extract_related_features(idea)\n    used_features = get_used_features(context['usage_log'])\n    \n    overlap_ratio = len(features & used_features) \u002F len(features)\n    score += int(overlap_ratio * 30)\n    \n    # Role fit\n    role = context['user_profile']['role']\n    category = idea['category']\n    \n    role_fit_map = {\n        'PM': ['projects', 'workflows', 'knowledge'],\n        'Sales': ['relationships', 'tasks'],\n        'Leadership': ['knowledge', 'workflows']\n    }\n    \n    if category in role_fit_map.get(role, []):\n        score += 20\n    \n    return min(score, 100)\n```\n\n### For Token Efficiency Calculation\n\n```python\ndef calculate_token_efficiency(idea, context):\n    score = 50\n    \n    # Check if reduces token usage\n    if reduces_reads(idea):  # Caching, summaries\n        score += 25\n    \n    # Context efficiency improvements\n    if improves_retrieval(idea):  # Better indexing, structured data\n        score += 15\n    \n    # Incremental updates\n    if supports_incremental(idea):  # Partial updates, lazy loading\n        score += 10\n    \n    return min(score, 100)\n```\n\n### For Memory & Learning Calculation\n\n```python\ndef calculate_memory_learning(idea, context):\n    score = 50\n    \n    # Memory persistence\n    if stores_learnings(idea):  # Knowledge base, historical context\n        score += 20\n    \n    # Self-learning capability\n    if enables_adaptation(idea):  # Learns from behavior, improves over time\n        score += 20\n    \n    # Feedback loops\n    if tracks_outcomes(idea):  # Measures effectiveness, refines\n        score += 10\n    \n    return min(score, 100)\n```\n\n### For Proactivity Calculation\n\n```python\ndef calculate_proactivity(idea, context):\n    score = 50\n    \n    # Anticipation capability\n    if enables_anticipation(idea):  # Surfaces info before asked\n        score += 25\n    \n    # Automation of routine decisions\n    if automates_decisions(idea):  # Handles repetitive choices\n        score += 15\n    \n    # Timing improvements\n    if improves_timing(idea):  # Right info at right time\n        score += 10\n    \n    return min(score, 100)\n```\n\n---\n\n## Best Practices\n\n1. **Run weekly** during `\u002Fweek-plan` or standalone\n2. **Don't obsess over scores** - they're guidance, not gospel\n3. **Trust your instinct** - high score + gut feel = go\n4. **Keep backlog lean** - max 20 active ideas\n5. **Archive implemented** - celebrate progress\n6. **Refine low scorers** - add detail to boost alignment\u002Fimpact\n\n---\n\n## Philosophy\n\nThe backlog isn't a todo list - it's a **decision support system**.\n\nScores help you:\n- Surface high-value work\n- Avoid shiny object syndrome\n- Align improvements with actual usage\n- Make intentional choices\n\nBut you're still the decision maker. If a low-scoring idea excites you, workshop it. The system serves you, not the other way around.\n\n---\n\n## Track Usage (Silent)\n\nUpdate `System\u002Fusage_log.md` to mark backlog review as used.\n\n**Analytics (Silent):**\n\nCall `track_event` with event_name `backlog_reviewed` and properties:\n- `ideas_count`\n\nThis only fires if the user has opted into analytics. No action needed if it returns \"analytics_disabled\".\n","## What This Command Does\n\n**In plain English:** AI-powered ranking of your Amp system improvement backlog based on current system state. Shows you what to build next.\n\n**When to use it:**\n- Weekly check-in on system improvements\n- After capturing several new ideas\n- When deciding what to work on next\n- During quarterly planning for system improvements\n\n**How to run it:**\n```\n\u002Famp-backlog              # Full review with re-ranking\n```\n\n---\n\n## Process Overview\n\n1. **Load context** - Read system state, usage patterns, learnings\n2. **Score ideas** - Calculate 5-dimension scores for each idea\n3. **Re-rank backlog** - Sort by weighted total score\n4. **Update file** - Write new rankings to `System\u002FAmp_Backlog.md`\n5. **Present top ideas** - Show top 5 with \"Why now?\" justification\n6. **Offer next steps** - Workshop, implement, or defer\n\n---\n\n## Step 1: Load System Context\n\nRead these files to understand current system state:\n\n### Required Files\n```\nSystem\u002FAmp_Backlog.md              # All ideas to score\nSystem\u002Fusage_log.md                # Feature adoption patterns\nSystem\u002Fuser-profile.yaml           # Role, preferences\nCLAUDE.md                          # Current capabilities\n```\n\n### Optional Files (if they exist)\n```\nSystem\u002FSession_Learnings\u002F           # Recent pain points (last 30 days)\n.claude\u002Fcommands\u002F                  # Available commands\ncore\u002Fmcp\u002F                          # MCP integrations\n06-Resources\u002FLearnings\u002F               # Captured patterns\n```\n\n### Extract Context\n\nBuild a context dictionary with:\n- **Usage patterns**: Which features are used vs. unused\n- **Role profile**: PM, Sales, Leadership, Engineer, etc.\n- **Pain points**: Recent friction from session learnings\n- **System capabilities**: What's currently available\n- **Backlog state**: All ideas and their current scores\n\n---\n\n## Step 2: Score Each Idea\n\nFor every active idea in the backlog, calculate 5 dimension scores.\n\n### ⚠️ CURSOR FEASIBILITY CHECK (Do This First!)\n\nBefore scoring ANY idea, validate it's actually implementable in Cursor:\n\n**What Cursor\u002FTerminal CAN do:**\n- ✅ Read and write files\n- ✅ Execute shell commands\n- ✅ Build MCP tools for structured operations\n- ✅ Parse and transform file contents\n- ✅ Create caches and indexes (file-based)\n- ✅ Run commands on schedules or triggers\n\n**What Cursor\u002FTerminal CANNOT do:**\n- ❌ Track user edits in real-time\n- ❌ Hook into Cursor internals\n- ❌ Monitor user actions passively\n- ❌ Access edit history without explicit file reads\n- ❌ Real-time background processes watching for changes\n\n**If idea requires something from the CANNOT list → Set all scores to 0 and flag as \"Not feasible in Cursor\"**\n\n---\n\nAfter feasibility check passes, score on 5 dimensions:\n\n### Dimension 1: Impact (35% weight)\n\n**Question:** How much would this improve daily workflow?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+20 if matches_recent_pain_points():\n  - Search System\u002FSession_Learnings\u002F for mentions of this issue\n  - Keywords from idea title\u002Fdescription appear in learnings\n  - Problem stated explicitly in recent notes\n\n+15 if affects_daily_workflow():\n  - Touches commands used >3x per week (from usage_log)\n  - Modifies core files (03-Tasks\u002FTasks.md, daily plans, person pages)\n  - Impacts repetitive actions\n\n+15 if has_compound_value():\n  - Enables other ideas in backlog\n  - Reduces technical debt\n  - Creates reusable patterns\n  - Unblocks multiple workflows\n\nMax: 100\n```\n\n**Examples:**\n- \"Auto-suggest person pages\": 95 (daily workflow + enables relationship tracking)\n- \"Export to blog\": 40 (nice-to-have, doesn't affect core workflow)\n\n---\n\n### Dimension 2: Alignment (20% weight)\n\n**Question:** Does this fit actual usage patterns?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+30 based on usage_overlap():\n  - Extract features idea depends on\n  - Check if those features are used (usage_log)\n  - Calculate overlap: (used_features \u002F total_features) * 30\n\n+20 if fits_role_profile():\n  - PM roles: prioritize project\u002Fproduct features\n  - Sales roles: prioritize relationship\u002Faccount features\n  - Leadership: prioritize synthesis\u002Freview features\n  - Match category to role focus areas\n\nMax: 100\n```\n\n**Examples:**\n- Idea needs \"person pages\" → Check if user has created person pages\n- If usage_log shows person pages = used → Higher alignment\n- If role = PM and idea = product features → +20 role fit\n\n---\n\n### Dimension 3: Token Efficiency (20% weight)\n\n**Question:** Does this reduce context\u002Ftoken usage?\n\n**CRITICAL - Cursor Feasibility Check:**\nBefore scoring, verify the idea is implementable in Cursor\u002FTerminal:\n- ✅ Can use: File read\u002Fwrite, MCP tools, command execution, file-based caching\n- ❌ Cannot use: Real-time edit tracking, Cursor internal hooks, monitoring user actions\n- If not feasible in Cursor → Score = 0 on all dimensions\n\n**Scoring logic:**\n```\nBase score: 50\n\n+25 if reduces_token_usage():\n  - Caches\u002Fstores frequently accessed data (in files\u002FMCP)\n  - Compresses or summarizes verbose content (file-based)\n  - Eliminates redundant reads\n  - Enables more efficient retrieval patterns\n\n+15 if improves_context_efficiency():\n  - Reduces number of files that need reading\n  - Creates structured summaries (YAML\u002FJSON files)\n  - Better indexing\u002Fsearch to avoid broad scans\n  - Moves data from markdown to structured format\n\n+10 if enables_incremental_updates():\n  - Supports partial updates instead of full rewrites\n  - Tracks changes in separate files\n  - Lazy loading or on-demand computation\n\nMax: 100\n```\n\n**Examples:**\n- \"Cache meeting summaries in YAML\": 90 (file-based, avoids re-reading)\n- \"Track user edits for learning\": 0 (NOT FEASIBLE - can't track edits)\n- \"Add new field to template\": 50 (neutral token impact)\n\n---\n\n### Dimension 4: Memory & Learning (15% weight)\n\n**Question:** Does this enhance system memory, persistence, or self-learning?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+20 if improves_memory_persistence():\n  - Stores learnings for future reference\n  - Creates retrievable knowledge base\n  - Captures patterns that compound over time\n  - Builds historical context\n\n+20 if enables_self_learning():\n  - System learns from user behavior\n  - Adapts recommendations based on patterns\n  - Builds preference models\n  - Improves predictions over time\n\n+10 if creates_feedback_loops():\n  - Tracks outcomes of suggestions\n  - Measures effectiveness of recommendations\n  - Refines based on what works\n\nMax: 100\n```\n\n**Examples:**\n- \"Learning pattern synthesizer\": 95 (captures + compounds knowledge)\n- \"Preference learning from edits\": 85 (system adapts over time)\n- \"Static template update\": 50 (no learning component)\n\n---\n\n### Dimension 5: Proactivity (10% weight)\n\n**Question:** Does this enable proactive concierge behavior?\n\n**Scoring logic:**\n```\nBase score: 50\n\n+25 if enables_anticipation():\n  - Surfaces relevant info before asked\n  - Predicts needs based on patterns\n  - Proactive suggestions not just reactive\n  - Context-aware prompts\n\n+15 if automates_routine_decisions():\n  - Handles repetitive choices automatically\n  - Learns user preferences and applies them\n  - Reduces decision fatigue\n\n+10 if improves_timing():\n  - Right information at right time\n  - Context-aware interruptions\n  - Anticipates upcoming needs\n\nMax: 100\n```\n\n**Examples:**\n- \"Auto-prep meetings based on calendar\": 90 (proactive + anticipatory)\n- \"Suggest weekly priorities from patterns\": 80 (learns and anticipates)\n- \"Add manual review step\": 50 (reactive, not proactive)\n\n---\n\n## Step 3: Calculate Weighted Score\n\n```\ntotal_score = (\n  (impact * 0.35) +\n  (alignment * 0.20) +\n  (token_efficiency * 0.20) +\n  (memory_learning * 0.15) +\n  (proactivity * 0.10)\n)\n\nRound to integer: total_score = round(total_score)\n```\n\n**Priority Bands:**\n- **High Priority (85+):** Should tackle soon, high ROI\n- **Medium Priority (60-84):** Good ideas, right time matters\n- **Low Priority (\u003C60):** Maybe later or needs refinement\n\n**Why These Dimensions:**\n- **Effort excluded:** With AI coding, implementation is cheap - focus on value, not cost\n- **Token efficiency prioritized:** Context efficiency is critical for performance\n- **Memory & learning emphasized:** System should get smarter over time\n- **Proactivity valued:** Concierge behavior > reactive tool\n\n---\n\n## Step 4: Update Backlog File\n\nRewrite `System\u002FAmp_Backlog.md` with:\n\n1. **Update timestamp** at top\n2. **Re-sort ideas** by total score (high to low)\n3. **Update each idea** with new scores:\n   ```markdown\n   - **[idea-XXX]** Title\n     - **Score:** 92 (Impact: 95, Alignment: 90, Effort: 85, Synergy: 95, Fresh: 70)\n     - **Category:** category\n     - **Captured:** YYYY-MM-DD\n     - **Why ranked here:** [1-2 sentence reasoning based on scores]\n     - **Description:** [original description]\n   ```\n4. **Place in correct section** (High\u002FMedium\u002FLow priority)\n5. **Preserve Archive section** (don't re-rank implemented ideas)\n\n---\n\n## Step 5: Present Results\n\nShow the user the top 5 ideas with context:\n\n```markdown\n# ? Backlog Review Complete\n\n*Analyzed {{total_ideas}} ideas against current system state*\n\n## ? Top 5 Recommendations\n\n### 1. [idea-XXX] {{title}} (Score: {{score}})\n\n**Why now:** {{reasoning based on scores - be specific}}\n\n**Quick assessment:**\n- Impact: {{impact_justification}}\n- Fits your patterns: {{alignment_justification}}\n- Effort: {{effort_estimate}}\n\n**Next step:** Run `\u002Famp-improve \"{{title}}\"` to workshop this idea\n\n---\n\n### 2. [idea-YYY] {{title}} (Score: {{score}})\n\n[Same format]\n\n---\n\n[... continue for top 5 ...]\n\n---\n\n## ? Backlog Health\n\n- **Total ideas:** {{total}}\n- **High priority (85+):** {{high_count}}\n- **Medium priority (60-84):** {{medium_count}}\n- **Low priority (\u003C60):** {{low_count}}\n\n{{#if high_count > 5}}\n⚠️ **Note:** You have {{high_count}} high-priority ideas. Consider tackling 1-2 this week to reduce backlog.\n{{\u002Fif}}\n\n{{#if low_count > 10}}\n? **Tip:** {{low_count}} low-priority ideas might be worth archiving or refining.\n{{\u002Fif}}\n\n---\n\n## What would you like to do?\n\n1. **Workshop an idea** → `\u002Famp-improve \"[title]\"`\n2. **Capture a new idea** → Use `capture_idea` MCP tool\n3. **Mark one implemented** → Use `mark_implemented` MCP tool\n4. **View full backlog** → Check `System\u002FAmp_Backlog.md`\n```\n\n---\n\n## Step 6: Handle Special Cases\n\n### If Backlog is Empty\n```markdown\n# ? Backlog Review\n\nYour backlog is empty! \n\nStart capturing improvement ideas:\n- Use the `capture_idea` MCP tool anytime you think \"I wish Amp did X\"\n- Run `\u002Famp-improve` to explore capability gaps\n- Run `\u002Famp-level-up` to discover unused features\n\nThe backlog system will help you track and prioritize ideas systematically.\n```\n\n### If No High Priority Ideas\n```markdown\n? **Good news:** No urgent improvements needed!\n\nYour system is working well. The backlog has ideas for later, but nothing critical right now.\n\nConsider:\n- Running `\u002Famp-level-up` to discover unused features\n- Capturing ideas as they come up\n- Reviewing backlog quarterly\n```\n\n### If Many Stale Ideas (>6 months old)\n```markdown\n⚠️ **Backlog maintenance needed**\n\nYou have {{stale_count}} ideas older than 6 months. These might be:\n- No longer relevant → Archive them\n- Still valuable but not urgent → Keep them\n- Worth revisiting with new context → Re-evaluate descriptions\n\nReview stale ideas:\n{{list stale ideas}}\n\nWant to bulk archive these? I can help clean up the backlog.\n```\n\n---\n\n## Integration with Other Commands\n\n### Hand-off to \u002Famp-improve\n\nWhen user says \"Let's work on #1\" or \"Workshop idea-XXX\":\n\n1. Read the idea details from backlog\n2. Pass to `\u002Famp-improve` with context:\n   ```\n   \u002Famp-improve \"{{idea_title}}\"\n   \n   Context from backlog:\n   - Current score: {{score}}\n   - Why it's prioritized: {{reasoning}}\n   - Original description: {{description}}\n   ```\n3. `\u002Famp-improve` takes over for workshopping\n\n---\n\n## Scoring Implementation Tips\n\n### Cursor Feasibility Check (Run FIRST)\n\n```python\ndef check_cursor_feasibility(idea: dict) -> dict:\n    \"\"\"\n    Returns: {\n        'feasible': bool,\n        'reason': str,\n        'capabilities_required': list\n    }\n    \"\"\"\n    description_lower = idea['description'].lower()\n    \n    # Red flags - things Cursor CAN'T do\n    cannot_do = {\n        'track edits': 'Cannot monitor file edits in real-time',\n        'watch user': 'Cannot observe user actions passively',\n        'hook into': 'Cannot hook into Cursor internals',\n        'monitor changes': 'Cannot monitor without explicit file reads',\n        'background process': 'No persistent background processes'\n    }\n    \n    for phrase, reason in cannot_do.items():\n        if phrase in description_lower:\n            return {\n                'feasible': False,\n                'reason': reason,\n                'suggestion': 'Reframe as file-based or command-triggered'\n            }\n    \n    # Green flags - things Cursor CAN do\n    can_do = ['file', 'read', 'write', 'mcp', 'command', 'cache', 'index', 'parse']\n    has_feasible_approach = any(word in description_lower for word in can_do)\n    \n    if has_feasible_approach:\n        return {'feasible': True, 'reason': 'Uses Cursor-compatible operations'}\n    else:\n        return {\n            'feasible': False,\n            'reason': 'No clear implementation path in Cursor',\n            'suggestion': 'Add file-based or MCP approach'\n        }\n```\n\n### For Impact Calculation\n\n```python\ndef calculate_impact(idea, context):\n    # First check feasibility\n    feasibility = check_cursor_feasibility(idea)\n    if not feasibility['feasible']:\n        return 0  # Not feasible = 0 impact\n    \n    score = 50\n    \n    # Check session learnings for pain point mentions\n    learnings = context['session_learnings']\n    idea_keywords = extract_keywords(idea['title'] + idea['description'])\n    \n    for learning in learnings:\n        learning_keywords = extract_keywords(learning['content'])\n        if overlap(idea_keywords, learning_keywords) > 0.3:\n            score += 20\n            break\n    \n    # Check if affects daily workflow\n    if touches_daily_commands(idea, context['usage_log']):\n        score += 15\n    \n    # Check compound value\n    if enables_other_ideas(idea, context['backlog']):\n        score += 15\n    \n    return min(score, 100)\n```\n\n### For Alignment Calculation\n\n```python\ndef calculate_alignment(idea, context):\n    score = 50\n    \n    # Extract related features\n    features = extract_related_features(idea)\n    used_features = get_used_features(context['usage_log'])\n    \n    overlap_ratio = len(features & used_features) \u002F len(features)\n    score += int(overlap_ratio * 30)\n    \n    # Role fit\n    role = context['user_profile']['role']\n    category = idea['category']\n    \n    role_fit_map = {\n        'PM': ['projects', 'workflows', 'knowledge'],\n        'Sales': ['relationships', 'tasks'],\n        'Leadership': ['knowledge', 'workflows']\n    }\n    \n    if category in role_fit_map.get(role, []):\n        score += 20\n    \n    return min(score, 100)\n```\n\n### For Token Efficiency Calculation\n\n```python\ndef calculate_token_efficiency(idea, context):\n    score = 50\n    \n    # Check if reduces token usage\n    if reduces_reads(idea):  # Caching, summaries\n        score += 25\n    \n    # Context efficiency improvements\n    if improves_retrieval(idea):  # Better indexing, structured data\n        score += 15\n    \n    # Incremental updates\n    if supports_incremental(idea):  # Partial updates, lazy loading\n        score += 10\n    \n    return min(score, 100)\n```\n\n### For Memory & Learning Calculation\n\n```python\ndef calculate_memory_learning(idea, context):\n    score = 50\n    \n    # Memory persistence\n    if stores_learnings(idea):  # Knowledge base, historical context\n        score += 20\n    \n    # Self-learning capability\n    if enables_adaptation(idea):  # Learns from behavior, improves over time\n        score += 20\n    \n    # Feedback loops\n    if tracks_outcomes(idea):  # Measures effectiveness, refines\n        score += 10\n    \n    return min(score, 100)\n```\n\n### For Proactivity Calculation\n\n```python\ndef calculate_proactivity(idea, context):\n    score = 50\n    \n    # Anticipation capability\n    if enables_anticipation(idea):  # Surfaces info before asked\n        score += 25\n    \n    # Automation of routine decisions\n    if automates_decisions(idea):  # Handles repetitive choices\n        score += 15\n    \n    # Timing improvements\n    if improves_timing(idea):  # Right info at right time\n        score += 10\n    \n    return min(score, 100)\n```\n\n---\n\n## Best Practices\n\n1. **Run weekly** during `\u002Fweek-plan` or standalone\n2. **Don't obsess over scores** - they're guidance, not gospel\n3. **Trust your instinct** - high score + gut feel = go\n4. **Keep backlog lean** - max 20 active ideas\n5. **Archive implemented** - celebrate progress\n6. **Refine low scorers** - add detail to boost alignment\u002Fimpact\n\n---\n\n## Philosophy\n\nThe backlog isn't a todo list - it's a **decision support system**.\n\nScores help you:\n- Surface high-value work\n- Avoid shiny object syndrome\n- Align improvements with actual usage\n- Make intentional choices\n\nBut you're still the decision maker. If a low-scoring idea excites you, workshop it. The system serves you, not the other way around.\n\n---\n\n## Track Usage (Silent)\n\nUpdate `System\u002Fusage_log.md` to mark backlog review as used.\n\n**Analytics (Silent):**\n\nCall `track_event` with event_name `backlog_reviewed` and properties:\n- `ideas_count`\n\nThis only fires if the user has opted into analytics. No action needed if it returns \"analytics_disabled\".\n",{"name":5,"description":7},"main","5417cad9215b767627c7c947dcb3a0bbe3ebe52d",".claude\u002Fskills\u002Famp-backlog",0,"https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use","https:\u002F\u002Fgithub.com\u002Fhemory","https:\u002F\u002Fwww.heminthehat.com\u002F","The People Developer | L&D Program Manager | Software Engineer","2026-06-04 23:46:30",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add hemory\u002Famp-legacy-do-not-use --skill amp-backlog",{"tool":43,"command":44},"bunx","bunx skills add hemory\u002Famp-legacy-do-not-use --skill amp-backlog",{"tool":46,"command":47},"pnpm","pnpm dlx skills add hemory\u002Famp-legacy-do-not-use --skill amp-backlog","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Famp-backlog\u002FSKILL.md",true,[],[52,89,98,112,127,139],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":68,"publishedAt":23},"01M6D3DVEAHJTTK4W21B15ZZKK","zoom-setup","Connect Zoom to Amp for meeting recordings, scheduling, and transcript context","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fzoom-setup\u002FSKILL.md",[58,63],{"categoryNo":59,"categoryName":60,"taxonomyType":61,"taxonomyTypeDesc":62},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",{"categoryNo":64,"categoryName":65,"taxonomyType":66,"taxonomyTypeDesc":67},"01JBSEEDCAT000000000000034","行政与办公人员","PROFESSION","适用职业",[69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88],"zoom setup","setup","schedule","MCP","review","connect","transcript","search","stats","oauth2","confirmation","context","meetings","process","recordings","surfaces","scheduling","granola","time","distribution",{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":23},"01M6D3DRAK8VWESYDZDF8JJDSH","xray","Understand what just happened under the hood - learn AI by seeing it in action","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fxray\u002FSKILL.md",[],[91,96,97],"learn","just",{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":105,"publishedAt":23},"01M6D3DN7BZQDY6143A2QSDP9A","week-review","Review week's progress with concrete accomplishments (not fake percentages), pattern detection, and goal tracking.","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fweek-review\u002FSKILL.md",[104],{"categoryNo":59,"categoryName":60,"taxonomyType":61,"taxonomyTypeDesc":62},[106,73,107,108,109,110,111],"week review","pattern detection","detection","tracking","goal","pattern",{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"publishedAt":23},"01M6D3DJ7F1TCFXBBSTQW82WE7","week-plan","Set weekly priorities and plan the week ahead with intelligent suggestions based on goals, calendar shape, and task effort.","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fweek-plan\u002FSKILL.md",[],[119,120,121,122,123,124,125,126],"week plan","plan","calendar","SET","shape","goals","task","priorities",{"skillNo":128,"name":129,"nameZh":6,"description":130,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":131,"categories":132,"keywords":133,"publishedAt":23},"01M6D3DF90D7RC8ZJE4AW2P9VC","triage","Strategically route orphaned files and extract scattered tasks","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftriage\u002FSKILL.md",[],[129,134,135,136,137,138],"tasks","fast","extract","files","route",{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":149,"publishedAt":23},"01M6D3DC9BBHK5TASE34SQYXSQ","trello-setup","Connect Trello to Amp for visual Kanban task sync","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftrello-setup\u002FSKILL.md",[145,148],{"categoryNo":146,"categoryName":147,"taxonomyType":61,"taxonomyTypeDesc":62},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":59,"categoryName":60,"taxonomyType":61,"taxonomyTypeDesc":62},[150,151,70,152,74,72,153,125,154,155,156,157,158],"trello","trello setup","sync","token","kanban","bun","visual","api","server"]