[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6D3ACJR6R3P9V58X8TE9J45":3,"skill-same:01M6D3ACJR6R3P9V58X8TE9J45":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},"01M6D3ACJR6R3P9V58X8TE9J45","getting-started",null,"Interactive post-onboarding tour with adaptive pathways based on available data","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\u002Fgetting-started\u002FSKILL.md",[],[19,20,21,22],"getting started","onboarding","data","interactive","2026-11-10 10:12:30",false,"---\nname: getting-started\ndescription: Interactive post-onboarding tour with adaptive pathways based on available data\n---\n\n## Purpose\n\nTransform the post-onboarding experience from \"blank chat window\" to guided value delivery. Adaptive based on what data sources are available (calendar, Granola, or neither).\n\n## When to Run\n\n- Automatically suggested at session start if vault \u003C 7 days old\n- User types `\u002Fgetting-started`\n- After onboarding completion\n- User says they're not sure what to do next\n\n## Entry Point\n\n### Step 1: Check for Deferred Pre-Analysis\n\n**FIRST:** Read `.onboarding-complete` marker file to check if this is Phase 2 right after onboarding:\n\n```python\nmarker_file = BASE_DIR \u002F 'System' \u002F '.onboarding-complete'\nif marker_file.exists():\n    marker_data = json.loads(marker_file.read_text())\n    pre_analysis_deferred = marker_data.get('pre_analysis_deferred', False)\n    completed_at = datetime.fromisoformat(marker_data['completed_at'])\n    age_hours = (datetime.now() - completed_at).total_seconds() \u002F 3600\n    \n    # If onboarding just completed (\u003C 1 hour) and pre-analysis was deferred\n    if age_hours \u003C 1 and pre_analysis_deferred:\n        # This is the first \u002Fgetting-started run after onboarding\n        # MCPs are NOW loaded and we can do the full analysis with dramatic reveal\n        run_dramatic_reveal = True\n    else:\n        run_dramatic_reveal = False\nelse:\n    run_dramatic_reveal = False\n```\n\n### Step 2: Dramatic Reveal (if flagged)\n\n**If `run_dramatic_reveal` is True:**\n\nSay: \"Hold on... analyzing your calendar and meetings. ?\"\n\n**Execute analysis:**\n1. Call calendar MCP: `get_events_for_week()` \n2. Call granola MCP: `get_recent_meetings(days=7)`\n3. Analyze the data\n\n**Then reveal what you found:**\n\n```\n**Here's what I found:**\n\n? **Your calendar this week:**\n• [X] meetings scheduled\n• [Y] are 1:1s  \n• [Busiest day] is packed ([Z] meetings back-to-back)\n• You're meeting most with: [Person1], [Person2], [Person3]\n\n[If Granola data:]\n? **Granola meetings (last 7 days):**\n• [N] meetings captured\n• [M] unique people\n• [K] external companies\n\n**I went ahead and created:**\n✅ Your weekly plan with time blocks around these meetings\n✅ Person pages for [Person1], [Person2], [Person3]\n✅ Identified [K] external organizations to track\n\nWant to see? [Show excerpt from 02-Week_Priorities\u002FWeek_Priorities.md]\n\n**This is what Amp does** - turns your calendar chaos into structure automatically.\n\nNow, want me to:\n1. Process those [N] Granola meetings (extract action items)?\n2. Create more person pages?\n3. Show you around the system?\n4. You're good - let me explore on my own\n\nWhat sounds useful?\n```\n\n**Important:** After showing this, update the marker file to prevent showing it again:\n```python\nmarker_data['phase2_completed'] = True\nmarker_data['phase2_completed_at'] = datetime.now().isoformat()\n```\n\n### Step 3: Standard Environment Check (if not flagged)\n\nCall `check_onboarding_complete()` from onboarding-mcp to verify vault status.\n\nThen check what data sources are available:\n- Calendar: Try calling calendar MCP (if fails, not available)\n- Granola: Call `granola_check_available()` from granola-mcp\n\nBased on results, route to appropriate flow.\n\n---\n\n## Flow A: Has Calendar + Granola\n\n**The \"I Already Looked\" Flow**\n\n### Pre-Check: Read Marker File\n\n**FIRST:** Check if `.onboarding-complete` marker file has `pre_analysis` data or deferred flag.\n\n```python\nmarker_file = BASE_DIR \u002F 'System' \u002F '.onboarding-complete'\nif marker_file.exists():\n    marker_data = json.loads(marker_file.read_text())\n    pre_analysis = marker_data.get('pre_analysis', {})\n    pre_analysis_deferred = marker_data.get('pre_analysis_deferred', False)\nelse:\n    pre_analysis = {}\n    pre_analysis_deferred = False\n```\n\n### If pre_analysis_deferred is True (first time running \u002Fgetting-started):\n\n**Run the full analysis NOW and do the dramatic reveal:**\n\n```\n? **Welcome, [Name]!**\n\nHold on - let me analyze what you have available...\n\n[Show spinner or progress indication]\n\n[Execute analysis:]\n1. Fetch calendar events for this week\n2. Analyze Granola meetings from last 7 days\n3. Identify top 3 frequent contacts\n4. Count unique people and external companies\n5. Create weekly plan with real data\n6. Create person pages for top contacts\n```\n\n**Then present the dramatic reveal:**\n\n```\n✓ **Analysis complete!**\n\nI just analyzed your data:\n\n**Calendar (this week):**\n• **[X] meetings total** [Add reaction: \"that's a lot!\" if >10, \"good balance\" if 5-10, \"light week\" if \u003C5]\n• **[Y] are 1:1s** [Add reaction based on role]\n• **[busiest_day] is your busiest day** ([Z] meetings)\n\n**Granola (last 7 days):**\n• **[N] meetings** with transcripts\n• **[M] unique people** \n• **[K] external companies**\n\n**I went ahead and:**\n✅ Created your weekly plan with time blocks based on these meetings\n✅ Made person pages for [Person1], [Person2], [Person3]\n✅ Identified [K] external organizations\n\nWant to see what I built? [Show excerpt from weekly plan]\n\n**This is what Amp does - it works while you're not looking.**\n\nNow, want me to process those [N] Granola meetings to extract action items and context?\n```\n\n**After dramatic reveal:**\n1. Update marker file: remove `pre_analysis_deferred`, add `pre_analysis` with the data\n2. Continue to backfill options\n\n### If pre_analysis exists (already ran \u002Fgetting-started before):\n\n```\n? **Welcome back, [Name]!**\n\nI already analyzed your data during setup. Let me refresh and see what's changed...\n\n[Check completed_at timestamp from marker]\n[Fetch new calendar events since that time]\n[Fetch new Granola meetings since that time]\n\nSince we set up [X hours\u002Fdays ago]:\n• [N] new calendar events\n• [M] new Granola meetings\n• [P] new people detected\n\n[If pre_analysis.created_weekly_plan:]\n✅ Your weekly plan is already created (based on [X] meetings)\n\n[If pre_analysis.created_person_pages:]\n✅ Person pages already created for [list names]\n\nWant me to:\n1. Process the [M] new Granola meetings?\n2. Create person pages for the [P] new people?\n3. Update your weekly plan with new meetings?\n\nOr everything looks current and you're ready to go?\n```\n\n### If no pre_analysis (legacy or failed):\n\n**Use standard \"I Already Looked\" flow:**\n\n### Discovery Phase\n\n```\n? **Welcome back, [Name]!**\n\nI just checked what you have available...\n\n**Calendar:** ✅ Connected\n**Granola:** ✅ Installed\n\nHold on - let me analyze what's there...\n```\n\nThen execute:\n1. Fetch calendar events for this week\n2. Analyze Granola data extent using helper function (6 months by default)\n3. If more data exists, ask if they want to see full extent\n4. Present comprehensive summary\n\n```\n? **Here's what I found:**\n\n**Calendar (this week):**\n• [X] meetings scheduled\n• [Y] are 1:1s\n• [Busiest day] is your busiest day ([Z] meetings)\n• Top contacts: [Person1], [Person2], [Person3]\n\n**Granola data (last 6 months):**\n• [N] meetings captured\n• Going back [days_back] days (to [oldest_date])\n• [M] unique people detected ([I] internal, [E] external)\n• [K] external companies identified\n```\n\n**If `has_more_data` is True, ask:**\n\n```\n? **I see there are meetings beyond 6 months.**\n   Want me to check how much more history you have?\n   This will take a few more seconds.\n   \n   → Yes - show me the full extent\n   → No - 6 months is plenty\n```\n\n**If user says \"Yes\", refetch with `extended=True` and update:**\n\n```\n? **Updated - Full data extent:**\n• [N_total] meetings captured (found [N_new] older meetings)\n• Going back [total_days] days (to [oldest_date])\n• [M_total] unique people detected\n• [K_total] external companies identified\n\n---\n\nHere's what I can create from your Granola history:\n\n**? People & Company Pages** (Recommended: All history)\n   ✅ Builds context for relationships\n   ✅ Low overhead - just reference pages\n   ✅ Helps you see interaction history\n\n**? Meeting Notes** (Recommended: Last 30 days)\n   ✅ Searchable record of discussions\n   ⚠️  Medium overhead - lots of reading material\n   ✅ Good for finding past decisions\n\n**✅ Action Items \u002F Todos** (Recommended: Last 7 days)\n   ✅ Actionable recent tasks\n   ⚠️  Can be overwhelming if too many\n   ⚠️  Old todos often outdated or already done\n\n**What would you like me to do?**\n\n1️⃣ Smart default (Recommended)\n   • People\u002Fcompanies: All [days_back] days\n   • Meeting notes: Last 30 days (~[est_30d] meetings)\n   • Todos: Last 7 days (~[est_7d] meetings)\n\n2️⃣ Recent only (Conservative)\n   • Everything: Last 7 days only\n\n3️⃣ Full history (Comprehensive)\n   • Everything: All [days_back] days\n\n4️⃣ Custom (You choose)\n   • Pick different time ranges for each type\n\n5️⃣ Just going forward\n   • Start fresh from today\n\n6️⃣ Skip for now\n```\n\n### User Chooses Processing Strategy\n\nUse AskUserQuestion tool. If AskUserQuestion is not available, prompt in CLI with the same numbered options and capture the selection:\n```json\n{\n  \"questions\": [{\n    \"id\": \"granola_strategy\",\n    \"prompt\": \"How would you like to process your Granola data?\",\n    \"allow_multiple\": false,\n    \"options\": [\n      {\"id\": \"smart\", \"label\": \"1️⃣ Smart default - People\u002Fcompanies (all) + Notes (30d) + Todos (7d)\"},\n      {\"id\": \"recent\", \"label\": \"2️⃣ Recent only - Everything from last 7 days\"},\n      {\"id\": \"full\", \"label\": \"3️⃣ Full history - Everything from all available data\"},\n      {\"id\": \"custom\", \"label\": \"4️⃣ Custom - I'll choose time ranges for each type\"},\n      {\"id\": \"forward\", \"label\": \"5️⃣ Just going forward - Start fresh from today\"},\n      {\"id\": \"skip\", \"label\": \"6️⃣ Skip for now\"}\n    ]\n  }]\n}\n```\n\n### Custom Mode Flow\n\nIf user selects \"custom\", ask for granular preferences:\n\n```json\n{\n  \"title\": \"Choose time ranges for historical data\",\n  \"questions\": [\n    {\n      \"id\": \"people_range\",\n      \"prompt\": \"People & Company Pages: How far back?\",\n      \"allow_multiple\": false,\n      \"options\": [\n        {\"id\": \"7d\", \"label\": \"Last 7 days\"},\n        {\"id\": \"30d\", \"label\": \"Last 30 days\"},\n        {\"id\": \"90d\", \"label\": \"Last 90 days\"},\n        {\"id\": \"all\", \"label\": \"All [days_back] days (recommended)\"},\n        {\"id\": \"none\", \"label\": \"Skip - don't create\"}\n      ]\n    },\n    {\n      \"id\": \"notes_range\",\n      \"prompt\": \"Meeting Notes: How far back?\",\n      \"allow_multiple\": false,\n      \"options\": [\n        {\"id\": \"7d\", \"label\": \"Last 7 days\"},\n        {\"id\": \"30d\", \"label\": \"Last 30 days (recommended)\"},\n        {\"id\": \"90d\", \"label\": \"Last 90 days\"},\n        {\"id\": \"all\", \"label\": \"All [days_back] days\"},\n        {\"id\": \"none\", \"label\": \"Skip - don't create\"}\n      ]\n    },\n    {\n      \"id\": \"todos_range\",\n      \"prompt\": \"Action Items \u002F Todos: How far back?\",\n      \"allow_multiple\": false,\n      \"options\": [\n        {\"id\": \"7d\", \"label\": \"Last 7 days (recommended)\"},\n        {\"id\": \"30d\", \"label\": \"Last 30 days\"},\n        {\"id\": \"90d\", \"label\": \"Last 90 days\"},\n        {\"id\": \"all\", \"label\": \"All [days_back] days\"},\n        {\"id\": \"none\", \"label\": \"Skip - don't create\"}\n      ]\n    }\n  ]\n}\n```\n\nMap responses to days:\n- \"7d\" → 7\n- \"30d\" → 30\n- \"90d\" → 90\n- \"all\" → [days_back from extent analysis]\n- \"none\" → 0 (skip)\n\n### Process Based on Selection\n\n**Map strategy to time ranges:**\n\n```python\ndef map_strategy_to_ranges(strategy: str, extent: dict) -> dict:\n    \"\"\"Convert user strategy choice to specific time ranges\"\"\"\n    \n    if strategy == \"smart\":\n        return {\n            'people_days': extent['days_back'],  # All history\n            'notes_days': min(30, extent['days_back']),  # Last 30 days or all if less\n            'todos_days': min(7, extent['days_back'])   # Last 7 days or all if less\n        }\n    elif strategy == \"recent\":\n        return {\n            'people_days': 7,\n            'notes_days': 7,\n            'todos_days': 7\n        }\n    elif strategy == \"full\":\n        return {\n            'people_days': extent['days_back'],\n            'notes_days': extent['days_back'],\n            'todos_days': extent['days_back']\n        }\n    elif strategy == \"forward\":\n        return {\n            'people_days': 0,\n            'notes_days': 0,\n            'todos_days': 0\n        }\n    elif strategy == \"skip\":\n        return None\n    # For \"custom\", ranges come from separate AskUserQuestion responses (or CLI fallback)\n```\n\n**Show confirmation before processing:**\n\n```\n**Here's what I'll do:**\n\n? People\u002FCompany Pages: Last [people_days] days\n   → ~[est_people] people pages, ~[est_companies] company pages\n\n? Meeting Notes: Last [notes_days] days\n   → ~[est_notes] meeting notes\n\n✅ Todos: Last [todos_days] days\n   → Estimated [est_todos] action items\n\n**This will take about 2-3 minutes. Ready?**\n```\n\n**Processing phases:**\n\n1. **Fetch meetings by maximum range needed:**\n   ```python\n   max_days = max(people_days, notes_days, todos_days)\n   all_meetings = granola_get_recent_meetings(days_back=max_days, limit=1000)\n   ```\n\n2. **Phase 1: People & Companies (if people_days > 0):**\n   ```python\n   # Filter meetings for people processing\n   people_meetings = [m for m in all_meetings \n                      if days_ago(m['date']) \u003C= people_days]\n   \n   # For each unique person:\n   # - Extract from all meetings in range\n   # - Create person page with meeting history\n   # - Route to Internal\u002F or External\u002F\n   # - Create\u002Fupdate company pages for external domains\n   \n   # Use \u002Fprocess-meetings --people-only --days-back={people_days} logic\n   ```\n\n3. **Phase 2: Meeting Notes (if notes_days > 0):**\n   ```python\n   # Filter meetings for note creation\n   notes_meetings = [m for m in all_meetings \n                     if days_ago(m['date']) \u003C= notes_days]\n   \n   # For each meeting:\n   # - Create detailed meeting note\n   # - Extract key points and decisions\n   # - Link to person\u002Fcompany pages\n   # - BUT don't extract todos yet\n   \n   # Use \u002Fprocess-meetings --no-todos --days-back={notes_days} logic\n   ```\n\n4. **Phase 3: Todos (if todos_days > 0):**\n   ```python\n   # Filter meetings for todo extraction\n   todo_meetings = [m for m in all_meetings \n                    if days_ago(m['date']) \u003C= todos_days]\n   \n   # For meetings that have notes:\n   # - Extract action items\n   # - Create tasks in 03-Tasks\u002FTasks.md\n   # - Add task IDs to meeting notes\n   \n   # For meetings in todo range but not notes range:\n   # - Still extract todos but note they came from unprocessed meetings\n   ```\n\n5. **Show completion summary:**\n   ```\n   ## Processing Complete ✅\n   \n   **People & Companies:**\n   • Created [X] person pages ([I] internal, [E] external)\n   • Created [Y] company pages\n   • Processed [P] days of history\n   \n   **Meeting Notes:**\n   • Created [N] detailed meeting notes\n   • From last [notes_days] days\n   \n   **Action Items:**\n   • Extracted [T] todos\n   • From last [todos_days] days\n   • Added to 03-Tasks\u002FTasks.md\n   \n   **Your vault now has:**\n   • Rich context from [people_days] days of meetings\n   • Searchable notes from [notes_days] days\n   • Actionable todos from [todos_days] days\n   \n   Want to explore what was created?\n   ```\n\n**If \"skip\" selected:**\n- Show: \"No problem! You can always run `\u002Fprocess-meetings` later when you're ready.\"\n- Move to completion flow\n\n**If \"forward\" selected:**\n- Show: \"Got it - starting fresh from today. I won't backfill historical data.\"\n- Update a marker file to remember this choice\n- Future meetings will be processed normally\n\n---\n\n## Flow B: Has Calendar OR Granola (Not Both)\n\n**The \"Let Me Help You Complete The Picture\" Flow**\n\n### If Only Calendar:\n\n```\n? **Welcome back, [Name]!**\n\nI can see your calendar for this week:\n• [X] meetings scheduled\n• [Y] are 1:1s\n• [Busiest day] looks intense\n\nI can create person pages for your frequent contacts if you'd like.\n\n**But I notice you don't have Granola** - that's how I process meeting transcripts into action items and insights.\n\nWant help with:\n1. Installing Granola (automatic meeting intelligence)\n2. Connecting another meeting tool\n3. Or tell me what other tools you use - I'll build integrations\n\nWhat sounds useful?\n```\n\n### If Only Granola:\n\n**Same discovery flow as Flow A, but calendar-less:**\n\n```\n? **Welcome back, [Name]!**\n\nI can see you have Granola installed. Let me check what's available...\n\n[Analyze Granola data extent - 6 months by default]\n\n? **Granola data (last 6 months):**\n• [N] meetings captured\n• Going back [days_back] days (to [oldest_date])\n• [M] unique people detected ([I] internal, [E] external)\n• [K] external companies identified\n```\n\n**If `has_more_data` is True:**\n\n```\n? **I see there are meetings beyond 6 months.**\n   Want me to check how much more history you have?\n   \n   → Yes - show me the full extent\n   → No - 6 months is plenty\n```\n\n**Then continue:**\n\n```\n---\n\nHere's what I can create from your Granola history:\n\n**? People & Company Pages** (Recommended: All history)\n   ✅ Builds context for relationships\n   ✅ Low overhead - just reference pages\n\n**? Meeting Notes** (Recommended: Last 30 days)\n   ✅ Searchable record of discussions\n   ⚠️  Medium overhead\n\n**✅ Action Items \u002F Todos** (Recommended: Last 7 days)\n   ✅ Actionable recent tasks\n   ⚠️  Can be overwhelming if too many\n\n**What would you like me to do?**\n\n1️⃣ Smart default - People\u002Fcompanies (all) + Notes (30d) + Todos (7d)\n2️⃣ Recent only - Everything from last 7 days\n3️⃣ Full history - Everything from all available data\n4️⃣ Custom - I'll choose time ranges for each type\n5️⃣ Just going forward - Start fresh from today\n6️⃣ Skip for now\n\n**I also notice your calendar isn't connected.** After processing, want to:\n• Connect Google Calendar (even if work is restricted)\n• Connect Apple Calendar\n• Or skip - I can work with just Granola\n```\n\nUse the same AskUserQuestion and processing logic as Flow A (or CLI fallback).\n\n---\n\n## Flow C: Neither Calendar Nor Granola\n\n**The \"Let's Connect Your Tools\" Flow**\n\n```\n? **Welcome back, [Name]!**\n\nYour workspace is set up, but you don't have calendar or meeting tools connected yet.\n\n**Let's get you integrated with the tools you actually use.**\n\nWhat tools do you use most? Examples:\n• Notion, Linear, Jira\n• Slack, Discord\n• GitHub, GitLab\n• Your company's internal tools\n• Newsletters, RSS feeds\n\n**Give me names or URLs** (URLs are better - just the root domain like \"notion.so\" or \"linear.app\")\n\nOnce you tell me, I'll:\n1. Check for API documentation\n2. Analyze what's possible\n3. Generate working MCP code\n4. Get you integrated in ~2 minutes\n\nWhat's your most important tool?\n```\n\n### Tool Integration Flow\n\nWhen user provides tool name\u002FURL:\n\n**Step 1: Find API docs**\n```python\nif url_provided:\n    # Fetch directly\n    base_url = normalize_url(user_input)\n    doc_content = web_fetch(base_url)\nelse:\n    # Search for docs\n    search_patterns = [\n        f\"{tool_name} API documentation\",\n        f\"{tool_name} developer docs\",\n        f\"{tool_name} API reference\"\n    ]\n    # Try WebSearch or direct fetch of common patterns\n    doc_url = find_api_docs(tool_name)\n    doc_content = web_fetch(doc_url)\n```\n\n**Step 2: Analyze capabilities**\n```\n\"Hold on - reading [Tool]'s API documentation...\"\n\n[Analyze doc_content for:]\n- Authentication methods\n- Available endpoints\n- Common operations\n- Rate limits\n\n\"Got it! I just read [Tool]'s API. Here's what we can build:\n\n**Possible integrations:**\n• [Capability 1] - [Amp use case]\n• [Capability 2] - [Amp use case]\n• [Capability 3] - [Amp use case]\n\nBased on your role ([role]), I'd focus on [specific capabilities].\n\nReady to build this? Takes about 2 minutes.\"\n```\n\n**Step 3: Generate MCP**\n\nFollow `\u002Fcreate-mcp` skill flow:\n1. Design tools based on API analysis\n2. Generate server code\n3. Configure authentication\n4. Test connection\n5. Update documentation\n\n**Step 4: The Magic Moment**\n```\n\"Your [Tool] integration is live! ✅\n\nTest it right now:\n'[Natural language query about that tool]'\n\nSee? Real data, not AI guessing. That's what MCP does.\"\n```\n\n**Step 5: Offer More**\n```\n\"Want to connect another tool? Or explore what else Amp can do?\"\n\n[If yes, repeat]\n[If no, show quick reference guide]\n```\n\n---\n\n## Additional Pathways (Available in Any Flow)\n\n### Google Workspace Setup\n\nIf user wants to connect Gmail\u002FCalendar:\n\n```\n\"Let's connect your Google account.\n\nThis works even if your work account has restrictions - we can use your personal Google if needed.\n\n**What we'll set up:**\n• Calendar sync (see meetings in Amp)\n• Gmail access (summarize newsletters, find emails)\n• Optional: Create daily digest\n\nI'll guide you through OAuth step-by-step.\n\nReady?\"\n```\n\nThen:\n1. Guide through OAuth flow for Google Calendar MCP\n2. Add Gmail MCP if they want\n3. Offer to set up newsletter digest\n\n### Information Diet Setup\n\nIf user mentions newsletters, RSS, or content consumption:\n\n```\n\"Want me to create a daily digest for you?\n\nI can pull from:\n• Newsletters (via Gmail)\n• YouTube channels you follow\n• RSS feeds from blogs\n• Specific websites you check regularly\n\nEach morning, you'd get:\n• Summaries of new content\n• Key points extracted\n• Novel insights highlighted\n• All in one place\n\nWhat sources matter most to you?\"\n```\n\nThen build appropriate scrapers\u002Fintegrations.\n\n### Smithery.ai Discovery\n\nAnytime during tool discussion:\n\n```\n\"By the way - there are 100+ pre-built MCP servers at **Smithery.ai**\n\nBrowse there for:\n• GitHub, Linear, Jira\n• Notion, Airtable, Sheets\n• Slack, Discord, Email\n• Databases, monitoring tools\n• And way more...\n\nFind something interesting? Just paste the URL here and I'll integrate it.\n\nOr we can build custom if you don't find what you need.\"\n```\n\n---\n\n## Setting Expectations\n\nThroughout ANY flow, include:\n\n```\n\"**Fair warning:** I can't promise everything will work perfectly.\n\nCompany restrictions, API limitations, weird auth flows - \nlots can go wrong in the real world.\n\n**But here's what's cool:** Even when it doesn't work, you learn:\n• How to use Claude to debug APIs\n• How to build integrations yourself\n• How MCP servers work\n• How to read API documentation\n\nIt's educational even when it fails.\n\nStill want to try?\"\n```\n\n---\n\n## Cursor UX Tip (If Running in Cursor)\n\nAfter a few file edits, offer:\n\n```\n\"**Quick Cursor tip:**\n\nYou're seeing 'Accept' prompts for each change I make. That's Cursor asking permission.\n\n**Useful at first** - you see what's happening\n**Annoying later** - slows things down\n\nIf you want to auto-accept:\n1. Settings (Cmd+,)\n2. Search \"always allow tool use\"\n3. Enable it\n\nYour choice - some prefer control, others prefer speed.\"\n```\n\nOnly show if:\n- Detected running in Cursor (check environment)\n- After 3-4 accept prompts shown\n\n---\n\n## Integration Discovery (Contextual)\n\n**Trigger:** Run this check if the vault is \u003C 7 days old AND few or no integrations are connected.\n\n### When to Check\n\nAfter completing the main pathway flow (A, B, or C) but before showing the completion message:\n\n1. Read `System\u002Fintegrations\u002Fconfig.yaml` (if it exists)\n2. Count how many integrations have `enabled: true` (exclude `slack` since it's a default)\n3. If connected integrations \u003C= 1 AND vault age \u003C 7 days, run the concierge\n\n### How to Run\n\nSearch the vault for references to tools the user hasn't connected yet (Slack, Todoist, Google Calendar, Zoom, etc.). Check file contents in `00-Inbox\u002F`, `04-Projects\u002F`, and `05-Areas\u002F` for mentions.\n\n### Presenting Recommendations\n\n**Only surface high-value items** (keep it light, this isn't onboarding, it's a nudge).\n\nFor each high_value integration found:\n\n```\nBy the way, I noticed you mention **[shortName]** in a few places\n([mentions] references across files like [first example file]).\n\nConnecting it would give you: [value proposition]\n\nSetup takes [setupTime]. Want to connect it now?\n```\n\n**If multiple high_value items:**\n\n```\nI also spotted signals for **[shortName2]** and **[shortName3]** in your notes.\nWant to connect any of these? Or run `\u002Fintegrate-mcp` anytime later.\n```\n\n**Rules:**\n- Maximum 2 integration suggestions (don't overwhelm)\n- Only mention high_value items (score >= 5) — skip moderate and available\n- If user says yes, run the setup skill inline then return to the getting-started completion\n- If user says no\u002Fskip\u002Flater, move on without pressure\n- Don't show this section if the user already went through integration setup during onboarding Step 8\n\n### Detecting Prior Setup\n\n\n---\n\n## Completion & Next Steps\n\nAfter any pathway completes:\n\n```\n\"**You're set up!**\n\n**Daily workflow:**\n• Run `\u002Fdaily-plan` each morning\n• Use `\u002Fmeeting-prep [person]` before meetings\n• Tell me about meetings - I'll extract action items\n\n**Discovery:**\n• `\u002Famp-level-up` - Find features you haven't tried\n• `\u002Fintegrate-mcp` - Add more tools anytime\n• Smithery.ai - Browse MCP marketplace\n\n**Come back anytime:**\n• `\u002Fgetting-started` - Run this tour again\n• Just ask in natural language - I'll figure out what you need\n\nWhat would you like to work on first?\"\n```\n\nUpdate marker file:\n```python\nmarker_data['phase2_completed'] = True\nmarker_data['phase2_completed_at'] = datetime.now().isoformat()\nmarker_data['pathways_completed'] = selected_pathways\n```\n\n---\n\n## Usage Log Integration\n\nBefore showing suggestions, check `System\u002Fusage_log.md`:\n\n```python\nusage = read_usage_log()\ntried_features = [f for f, checked in usage.items() if checked]\n\n# Adaptive suggestions based on what they've used\nif len(tried_features) \u003C 3:\n    # New user - basics\n    suggest = [\"daily-plan\", \"meeting-prep\"]\nelif \"meeting-prep\" not in tried_features and has_calendar:\n    # Have meetings but haven't prepped\n    suggest = [\"meeting-prep\"]\nelif \"process-meetings\" not in tried_features and has_granola:\n    # Have Granola but haven't processed\n    suggest = [\"process-meetings\"]\nelse:\n    # Power user - advanced features\n    suggest = get_advanced_features(usage, role)\n```\n\nDon't show features they've already tried - focus on gaps.\n\n---\n\n## Escape Hatches\n\nAt EVERY major decision point:\n\n- \"This is a lot - want to pause and come back later?\"\n- \"No pressure - you can always run `\u002Fgetting-started` again\"\n- \"Want to explore on your own first? That's cool\"\n\nAfter full tour:\n- \"Remember - invoke `\u002Famp-level-up` to discover more\"\n- \"Or `\u002Fintegrate-mcp` to add tools as you need them\"\n- \"The system grows with you\"\n\n---\n\n## Success Metric\n\nThe moment they think: **\"How did it know to do that?\"**\n\nDifferent magic for different situations:\n- **With data:** Analyzed calendar\u002FGranola intelligently, offered smart actions\n- **Without data:** Built working integration in 2 minutes from just a name\n- **Either way:** They see value immediately and know where to go next\n\n---\n\n## Helper: Analyze Granola Data Extent\n\nBefore presenting choices, discover how much Granola data is available:\n\n```python\ndef analyze_granola_extent(user_email_domain: str, extended: bool = False) -> dict:\n    \"\"\"\n    Discover how much Granola data is available\n    \n    Args:\n        user_email_domain: User's company email domain for internal\u002Fexternal classification\n        extended: If True, fetch up to 2 years. If False, fetch 6 months (default)\n    \n    Returns:\n        Dictionary with extent analysis or None if no data\n    \"\"\"\n    from datetime import datetime\n    \n    # Default to 6 months for speed, optionally extend to 2 years\n    days_to_fetch = 365 * 2 if extended else 180\n    result = granola_get_recent_meetings(days_back=days_to_fetch, limit=1000)\n    \n    if not result.get('success') or not result.get('meetings'):\n        return {\n            'meetings_count': 0,\n            'days_back': 0,\n            'has_data': False\n        }\n    \n    meetings = result['meetings']\n    \n    # Find oldest and newest dates\n    dates = [m['date'] for m in meetings if m.get('date')]\n    if not dates:\n        return {'meetings_count': 0, 'days_back': 0, 'has_data': False}\n    \n    oldest = min(dates)\n    newest = max(dates)\n    \n    # Calculate days back\n    oldest_dt = datetime.fromisoformat(oldest)\n    newest_dt = datetime.fromisoformat(newest)\n    days_back = (newest_dt - oldest_dt).days + 1  # +1 to include both days\n    \n    # Extract unique people and companies\n    people = set()\n    internal_people = set()\n    external_people = set()\n    companies = set()\n    \n    # Normalize user domain for comparison\n    user_domains = [d.strip().lower() for d in user_email_domain.split(',')]\n    \n    for meeting in meetings:\n        for participant in meeting.get('participants', []):\n            name = participant.get('name')\n            email = participant.get('email')\n            \n            if name:\n                people.add(name)\n                \n                if email:\n                    domain = email.split('@')[1].lower() if '@' in email else None\n                    \n                    # Classify as internal or external\n                    if domain and any(d in domain or domain in d for d in user_domains):\n                        internal_people.add(name)\n                    else:\n                        external_people.add(name)\n                        if domain:\n                            companies.add(domain)\n                else:\n                    # No email provided - default to external\n                    external_people.add(name)\n    \n    # Calculate meetings in different time ranges for estimation\n    now = datetime.now()\n    meetings_7d = sum(1 for m in meetings if m.get('date') and \n                      (now - datetime.fromisoformat(m['date'])).days \u003C= 7)\n    meetings_30d = sum(1 for m in meetings if m.get('date') and \n                       (now - datetime.fromisoformat(m['date'])).days \u003C= 30)\n    meetings_90d = sum(1 for m in meetings if m.get('date') and \n                       (now - datetime.fromisoformat(m['date'])).days \u003C= 90)\n    \n    # Check if there might be more data beyond what we fetched\n    has_more = False\n    if not extended and len(meetings) >= 900:  # Close to limit, likely more data\n        has_more = True\n    \n    # Check if oldest meeting is exactly at the boundary (suggests more data)\n    if not extended and days_back >= 175:  # Close to 180 days\n        has_more = True\n    \n    return {\n        'has_data': True,\n        'meetings_count': len(meetings),\n        'days_back': days_back,\n        'oldest_date': oldest,\n        'newest_date': newest,\n        'unique_people': len(people),\n        'internal_people': len(internal_people),\n        'external_people': len(external_people),\n        'unique_companies': len(companies),\n        'people_sample': list(people)[:10],\n        'companies_list': list(companies),\n        'meetings_7d': meetings_7d,\n        'meetings_30d': meetings_30d,\n        'meetings_90d': meetings_90d,\n        'has_more_data': has_more,\n        'fetched_range_days': days_to_fetch\n    }\n```\n\n**Usage in Flow A:**\n```python\n# After detecting Granola is available\nuser_profile = read_user_profile()\nemail_domain = user_profile.get('email_domain', '')\n\n# Fetch initial 6 months\nextent = analyze_granola_extent(email_domain)\n\nif not extent['has_data']:\n    # Handle no data case\n    pass\nelse:\n    # Check if there's more data beyond 6 months\n    if extent['has_more_data']:\n        # Ask if they want to fetch more\n        response = AskUserQuestion({\n            \"questions\": [{\n                \"id\": \"fetch_more\",\n                \"prompt\": f\"I found {extent['meetings_count']} meetings going back {extent['days_back']} days. There appears to be more data beyond that. Want me to check how much more?\",\n                \"allow_multiple\": false,\n                \"options\": [\n                    {\"id\": \"yes\", \"label\": \"Yes - show me the full extent\"},\n                    {\"id\": \"no\", \"label\": \"No - 6 months is enough\"}\n                ]\n            }]\n        })\n        \n        if response['fetch_more'] == 'yes':\n            # Fetch extended range (2 years)\n            extent = analyze_granola_extent(email_domain, extended=True)\n    \n    # Show discovery summary and choice framework\n    pass\n```\n\n---\n\n## Track Usage (Silent)\n\nUpdate `System\u002Fusage_log.md` to mark getting started as used.\n\n**Analytics (Silent):**\n\nCall `track_event` with event_name `getting_started_completed` and properties:\n- calendar_connected\n- granola_connected\n- artifacts_created\n\nThis only fires if the user has opted into analytics. No action needed if it returns \"analytics_disabled\".\n","## Purpose\n\nTransform the post-onboarding experience from \"blank chat window\" to guided value delivery. Adaptive based on what data sources are available (calendar, Granola, or neither).\n\n## When to Run\n\n- Automatically suggested at session start if vault \u003C 7 days old\n- User types `\u002Fgetting-started`\n- After onboarding completion\n- User says they're not sure what to do next\n\n## Entry Point\n\n### Step 1: Check for Deferred Pre-Analysis\n\n**FIRST:** Read `.onboarding-complete` marker file to check if this is Phase 2 right after onboarding:\n\n```python\nmarker_file = BASE_DIR \u002F 'System' \u002F '.onboarding-complete'\nif marker_file.exists():\n    marker_data = json.loads(marker_file.read_text())\n    pre_analysis_deferred = marker_data.get('pre_analysis_deferred', False)\n    completed_at = datetime.fromisoformat(marker_data['completed_at'])\n    age_hours = (datetime.now() - completed_at).total_seconds() \u002F 3600\n    \n    # If onboarding just completed (\u003C 1 hour) and pre-analysis was deferred\n    if age_hours \u003C 1 and pre_analysis_deferred:\n        # This is the first \u002Fgetting-started run after onboarding\n        # MCPs are NOW loaded and we can do the full analysis with dramatic reveal\n        run_dramatic_reveal = True\n    else:\n        run_dramatic_reveal = False\nelse:\n    run_dramatic_reveal = False\n```\n\n### Step 2: Dramatic Reveal (if flagged)\n\n**If `run_dramatic_reveal` is True:**\n\nSay: \"Hold on... analyzing your calendar and meetings. ?\"\n\n**Execute analysis:**\n1. Call calendar MCP: `get_events_for_week()` \n2. Call granola MCP: `get_recent_meetings(days=7)`\n3. Analyze the data\n\n**Then reveal what you found:**\n\n```\n**Here's what I found:**\n\n? **Your calendar this week:**\n• [X] meetings scheduled\n• [Y] are 1:1s  \n• [Busiest day] is packed ([Z] meetings back-to-back)\n• You're meeting most with: [Person1], [Person2], [Person3]\n\n[If Granola data:]\n? **Granola meetings (last 7 days):**\n• [N] meetings captured\n• [M] unique people\n• [K] external companies\n\n**I went ahead and created:**\n✅ Your weekly plan with time blocks around these meetings\n✅ Person pages for [Person1], [Person2], [Person3]\n✅ Identified [K] external organizations to track\n\nWant to see? [Show excerpt from 02-Week_Priorities\u002FWeek_Priorities.md]\n\n**This is what Amp does** - turns your calendar chaos into structure automatically.\n\nNow, want me to:\n1. Process those [N] Granola meetings (extract action items)?\n2. Create more person pages?\n3. Show you around the system?\n4. You're good - let me explore on my own\n\nWhat sounds useful?\n```\n\n**Important:** After showing this, update the marker file to prevent showing it again:\n```python\nmarker_data['phase2_completed'] = True\nmarker_data['phase2_completed_at'] = datetime.now().isoformat()\n```\n\n### Step 3: Standard Environment Check (if not flagged)\n\nCall `check_onboarding_complete()` from onboarding-mcp to verify vault status.\n\nThen check what data sources are available:\n- Calendar: Try calling calendar MCP (if fails, not available)\n- Granola: Call `granola_check_available()` from granola-mcp\n\nBased on results, route to appropriate flow.\n\n---\n\n## Flow A: Has Calendar + Granola\n\n**The \"I Already Looked\" Flow**\n\n### Pre-Check: Read Marker File\n\n**FIRST:** Check if `.onboarding-complete` marker file has `pre_analysis` data or deferred flag.\n\n```python\nmarker_file = BASE_DIR \u002F 'System' \u002F '.onboarding-complete'\nif marker_file.exists():\n    marker_data = json.loads(marker_file.read_text())\n    pre_analysis = marker_data.get('pre_analysis', {})\n    pre_analysis_deferred = marker_data.get('pre_analysis_deferred', False)\nelse:\n    pre_analysis = {}\n    pre_analysis_deferred = False\n```\n\n### If pre_analysis_deferred is True (first time running \u002Fgetting-started):\n\n**Run the full analysis NOW and do the dramatic reveal:**\n\n```\n? **Welcome, [Name]!**\n\nHold on - let me analyze what you have available...\n\n[Show spinner or progress indication]\n\n[Execute analysis:]\n1. Fetch calendar events for this week\n2. Analyze Granola meetings from last 7 days\n3. Identify top 3 frequent contacts\n4. Count unique people and external companies\n5. Create weekly plan with real data\n6. Create person pages for top contacts\n```\n\n**Then present the dramatic reveal:**\n\n```\n✓ **Analysis complete!**\n\nI just analyzed your data:\n\n**Calendar (this week):**\n• **[X] meetings total** [Add reaction: \"that's a lot!\" if >10, \"good balance\" if 5-10, \"light week\" if \u003C5]\n• **[Y] are 1:1s** [Add reaction based on role]\n• **[busiest_day] is your busiest day** ([Z] meetings)\n\n**Granola (last 7 days):**\n• **[N] meetings** with transcripts\n• **[M] unique people** \n• **[K] external companies**\n\n**I went ahead and:**\n✅ Created your weekly plan with time blocks based on these meetings\n✅ Made person pages for [Person1], [Person2], [Person3]\n✅ Identified [K] external organizations\n\nWant to see what I built? [Show excerpt from weekly plan]\n\n**This is what Amp does - it works while you're not looking.**\n\nNow, want me to process those [N] Granola meetings to extract action items and context?\n```\n\n**After dramatic reveal:**\n1. Update marker file: remove `pre_analysis_deferred`, add `pre_analysis` with the data\n2. Continue to backfill options\n\n### If pre_analysis exists (already ran \u002Fgetting-started before):\n\n```\n? **Welcome back, [Name]!**\n\nI already analyzed your data during setup. Let me refresh and see what's changed...\n\n[Check completed_at timestamp from marker]\n[Fetch new calendar events since that time]\n[Fetch new Granola meetings since that time]\n\nSince we set up [X hours\u002Fdays ago]:\n• [N] new calendar events\n• [M] new Granola meetings\n• [P] new people detected\n\n[If pre_analysis.created_weekly_plan:]\n✅ Your weekly plan is already created (based on [X] meetings)\n\n[If pre_analysis.created_person_pages:]\n✅ Person pages already created for [list names]\n\nWant me to:\n1. Process the [M] new Granola meetings?\n2. Create person pages for the [P] new people?\n3. Update your weekly plan with new meetings?\n\nOr everything looks current and you're ready to go?\n```\n\n### If no pre_analysis (legacy or failed):\n\n**Use standard \"I Already Looked\" flow:**\n\n### Discovery Phase\n\n```\n? **Welcome back, [Name]!**\n\nI just checked what you have available...\n\n**Calendar:** ✅ Connected\n**Granola:** ✅ Installed\n\nHold on - let me analyze what's there...\n```\n\nThen execute:\n1. Fetch calendar events for this week\n2. Analyze Granola data extent using helper function (6 months by default)\n3. If more data exists, ask if they want to see full extent\n4. Present comprehensive summary\n\n```\n? **Here's what I found:**\n\n**Calendar (this week):**\n• [X] meetings scheduled\n• [Y] are 1:1s\n• [Busiest day] is your busiest day ([Z] meetings)\n• Top contacts: [Person1], [Person2], [Person3]\n\n**Granola data (last 6 months):**\n• [N] meetings captured\n• Going back [days_back] days (to [oldest_date])\n• [M] unique people detected ([I] internal, [E] external)\n• [K] external companies identified\n```\n\n**If `has_more_data` is True, ask:**\n\n```\n? **I see there are meetings beyond 6 months.**\n   Want me to check how much more history you have?\n   This will take a few more seconds.\n   \n   → Yes - show me the full extent\n   → No - 6 months is plenty\n```\n\n**If user says \"Yes\", refetch with `extended=True` and update:**\n\n```\n? **Updated - Full data extent:**\n• [N_total] meetings captured (found [N_new] older meetings)\n• Going back [total_days] days (to [oldest_date])\n• [M_total] unique people detected\n• [K_total] external companies identified\n\n---\n\nHere's what I can create from your Granola history:\n\n**? People & Company Pages** (Recommended: All history)\n   ✅ Builds context for relationships\n   ✅ Low overhead - just reference pages\n   ✅ Helps you see interaction history\n\n**? Meeting Notes** (Recommended: Last 30 days)\n   ✅ Searchable record of discussions\n   ⚠️  Medium overhead - lots of reading material\n   ✅ Good for finding past decisions\n\n**✅ Action Items \u002F Todos** (Recommended: Last 7 days)\n   ✅ Actionable recent tasks\n   ⚠️  Can be overwhelming if too many\n   ⚠️  Old todos often outdated or already done\n\n**What would you like me to do?**\n\n1️⃣ Smart default (Recommended)\n   • People\u002Fcompanies: All [days_back] days\n   • Meeting notes: Last 30 days (~[est_30d] meetings)\n   • Todos: Last 7 days (~[est_7d] meetings)\n\n2️⃣ Recent only (Conservative)\n   • Everything: Last 7 days only\n\n3️⃣ Full history (Comprehensive)\n   • Everything: All [days_back] days\n\n4️⃣ Custom (You choose)\n   • Pick different time ranges for each type\n\n5️⃣ Just going forward\n   • Start fresh from today\n\n6️⃣ Skip for now\n```\n\n### User Chooses Processing Strategy\n\nUse AskUserQuestion tool. If AskUserQuestion is not available, prompt in CLI with the same numbered options and capture the selection:\n```json\n{\n  \"questions\": [{\n    \"id\": \"granola_strategy\",\n    \"prompt\": \"How would you like to process your Granola data?\",\n    \"allow_multiple\": false,\n    \"options\": [\n      {\"id\": \"smart\", \"label\": \"1️⃣ Smart default - People\u002Fcompanies (all) + Notes (30d) + Todos (7d)\"},\n      {\"id\": \"recent\", \"label\": \"2️⃣ Recent only - Everything from last 7 days\"},\n      {\"id\": \"full\", \"label\": \"3️⃣ Full history - Everything from all available data\"},\n      {\"id\": \"custom\", \"label\": \"4️⃣ Custom - I'll choose time ranges for each type\"},\n      {\"id\": \"forward\", \"label\": \"5️⃣ Just going forward - Start fresh from today\"},\n      {\"id\": \"skip\", \"label\": \"6️⃣ Skip for now\"}\n    ]\n  }]\n}\n```\n\n### Custom Mode Flow\n\nIf user selects \"custom\", ask for granular preferences:\n\n```json\n{\n  \"title\": \"Choose time ranges for historical data\",\n  \"questions\": [\n    {\n      \"id\": \"people_range\",\n      \"prompt\": \"People & Company Pages: How far back?\",\n      \"allow_multiple\": false,\n      \"options\": [\n        {\"id\": \"7d\", \"label\": \"Last 7 days\"},\n        {\"id\": \"30d\", \"label\": \"Last 30 days\"},\n        {\"id\": \"90d\", \"label\": \"Last 90 days\"},\n        {\"id\": \"all\", \"label\": \"All [days_back] days (recommended)\"},\n        {\"id\": \"none\", \"label\": \"Skip - don't create\"}\n      ]\n    },\n    {\n      \"id\": \"notes_range\",\n      \"prompt\": \"Meeting Notes: How far back?\",\n      \"allow_multiple\": false,\n      \"options\": [\n        {\"id\": \"7d\", \"label\": \"Last 7 days\"},\n        {\"id\": \"30d\", \"label\": \"Last 30 days (recommended)\"},\n        {\"id\": \"90d\", \"label\": \"Last 90 days\"},\n        {\"id\": \"all\", \"label\": \"All [days_back] days\"},\n        {\"id\": \"none\", \"label\": \"Skip - don't create\"}\n      ]\n    },\n    {\n      \"id\": \"todos_range\",\n      \"prompt\": \"Action Items \u002F Todos: How far back?\",\n      \"allow_multiple\": false,\n      \"options\": [\n        {\"id\": \"7d\", \"label\": \"Last 7 days (recommended)\"},\n        {\"id\": \"30d\", \"label\": \"Last 30 days\"},\n        {\"id\": \"90d\", \"label\": \"Last 90 days\"},\n        {\"id\": \"all\", \"label\": \"All [days_back] days\"},\n        {\"id\": \"none\", \"label\": \"Skip - don't create\"}\n      ]\n    }\n  ]\n}\n```\n\nMap responses to days:\n- \"7d\" → 7\n- \"30d\" → 30\n- \"90d\" → 90\n- \"all\" → [days_back from extent analysis]\n- \"none\" → 0 (skip)\n\n### Process Based on Selection\n\n**Map strategy to time ranges:**\n\n```python\ndef map_strategy_to_ranges(strategy: str, extent: dict) -> dict:\n    \"\"\"Convert user strategy choice to specific time ranges\"\"\"\n    \n    if strategy == \"smart\":\n        return {\n            'people_days': extent['days_back'],  # All history\n            'notes_days': min(30, extent['days_back']),  # Last 30 days or all if less\n            'todos_days': min(7, extent['days_back'])   # Last 7 days or all if less\n        }\n    elif strategy == \"recent\":\n        return {\n            'people_days': 7,\n            'notes_days': 7,\n            'todos_days': 7\n        }\n    elif strategy == \"full\":\n        return {\n            'people_days': extent['days_back'],\n            'notes_days': extent['days_back'],\n            'todos_days': extent['days_back']\n        }\n    elif strategy == \"forward\":\n        return {\n            'people_days': 0,\n            'notes_days': 0,\n            'todos_days': 0\n        }\n    elif strategy == \"skip\":\n        return None\n    # For \"custom\", ranges come from separate AskUserQuestion responses (or CLI fallback)\n```\n\n**Show confirmation before processing:**\n\n```\n**Here's what I'll do:**\n\n? People\u002FCompany Pages: Last [people_days] days\n   → ~[est_people] people pages, ~[est_companies] company pages\n\n? Meeting Notes: Last [notes_days] days\n   → ~[est_notes] meeting notes\n\n✅ Todos: Last [todos_days] days\n   → Estimated [est_todos] action items\n\n**This will take about 2-3 minutes. Ready?**\n```\n\n**Processing phases:**\n\n1. **Fetch meetings by maximum range needed:**\n   ```python\n   max_days = max(people_days, notes_days, todos_days)\n   all_meetings = granola_get_recent_meetings(days_back=max_days, limit=1000)\n   ```\n\n2. **Phase 1: People & Companies (if people_days > 0):**\n   ```python\n   # Filter meetings for people processing\n   people_meetings = [m for m in all_meetings \n                      if days_ago(m['date']) \u003C= people_days]\n   \n   # For each unique person:\n   # - Extract from all meetings in range\n   # - Create person page with meeting history\n   # - Route to Internal\u002F or External\u002F\n   # - Create\u002Fupdate company pages for external domains\n   \n   # Use \u002Fprocess-meetings --people-only --days-back={people_days} logic\n   ```\n\n3. **Phase 2: Meeting Notes (if notes_days > 0):**\n   ```python\n   # Filter meetings for note creation\n   notes_meetings = [m for m in all_meetings \n                     if days_ago(m['date']) \u003C= notes_days]\n   \n   # For each meeting:\n   # - Create detailed meeting note\n   # - Extract key points and decisions\n   # - Link to person\u002Fcompany pages\n   # - BUT don't extract todos yet\n   \n   # Use \u002Fprocess-meetings --no-todos --days-back={notes_days} logic\n   ```\n\n4. **Phase 3: Todos (if todos_days > 0):**\n   ```python\n   # Filter meetings for todo extraction\n   todo_meetings = [m for m in all_meetings \n                    if days_ago(m['date']) \u003C= todos_days]\n   \n   # For meetings that have notes:\n   # - Extract action items\n   # - Create tasks in 03-Tasks\u002FTasks.md\n   # - Add task IDs to meeting notes\n   \n   # For meetings in todo range but not notes range:\n   # - Still extract todos but note they came from unprocessed meetings\n   ```\n\n5. **Show completion summary:**\n   ```\n   ## Processing Complete ✅\n   \n   **People & Companies:**\n   • Created [X] person pages ([I] internal, [E] external)\n   • Created [Y] company pages\n   • Processed [P] days of history\n   \n   **Meeting Notes:**\n   • Created [N] detailed meeting notes\n   • From last [notes_days] days\n   \n   **Action Items:**\n   • Extracted [T] todos\n   • From last [todos_days] days\n   • Added to 03-Tasks\u002FTasks.md\n   \n   **Your vault now has:**\n   • Rich context from [people_days] days of meetings\n   • Searchable notes from [notes_days] days\n   • Actionable todos from [todos_days] days\n   \n   Want to explore what was created?\n   ```\n\n**If \"skip\" selected:**\n- Show: \"No problem! You can always run `\u002Fprocess-meetings` later when you're ready.\"\n- Move to completion flow\n\n**If \"forward\" selected:**\n- Show: \"Got it - starting fresh from today. I won't backfill historical data.\"\n- Update a marker file to remember this choice\n- Future meetings will be processed normally\n\n---\n\n## Flow B: Has Calendar OR Granola (Not Both)\n\n**The \"Let Me Help You Complete The Picture\" Flow**\n\n### If Only Calendar:\n\n```\n? **Welcome back, [Name]!**\n\nI can see your calendar for this week:\n• [X] meetings scheduled\n• [Y] are 1:1s\n• [Busiest day] looks intense\n\nI can create person pages for your frequent contacts if you'd like.\n\n**But I notice you don't have Granola** - that's how I process meeting transcripts into action items and insights.\n\nWant help with:\n1. Installing Granola (automatic meeting intelligence)\n2. Connecting another meeting tool\n3. Or tell me what other tools you use - I'll build integrations\n\nWhat sounds useful?\n```\n\n### If Only Granola:\n\n**Same discovery flow as Flow A, but calendar-less:**\n\n```\n? **Welcome back, [Name]!**\n\nI can see you have Granola installed. Let me check what's available...\n\n[Analyze Granola data extent - 6 months by default]\n\n? **Granola data (last 6 months):**\n• [N] meetings captured\n• Going back [days_back] days (to [oldest_date])\n• [M] unique people detected ([I] internal, [E] external)\n• [K] external companies identified\n```\n\n**If `has_more_data` is True:**\n\n```\n? **I see there are meetings beyond 6 months.**\n   Want me to check how much more history you have?\n   \n   → Yes - show me the full extent\n   → No - 6 months is plenty\n```\n\n**Then continue:**\n\n```\n---\n\nHere's what I can create from your Granola history:\n\n**? People & Company Pages** (Recommended: All history)\n   ✅ Builds context for relationships\n   ✅ Low overhead - just reference pages\n\n**? Meeting Notes** (Recommended: Last 30 days)\n   ✅ Searchable record of discussions\n   ⚠️  Medium overhead\n\n**✅ Action Items \u002F Todos** (Recommended: Last 7 days)\n   ✅ Actionable recent tasks\n   ⚠️  Can be overwhelming if too many\n\n**What would you like me to do?**\n\n1️⃣ Smart default - People\u002Fcompanies (all) + Notes (30d) + Todos (7d)\n2️⃣ Recent only - Everything from last 7 days\n3️⃣ Full history - Everything from all available data\n4️⃣ Custom - I'll choose time ranges for each type\n5️⃣ Just going forward - Start fresh from today\n6️⃣ Skip for now\n\n**I also notice your calendar isn't connected.** After processing, want to:\n• Connect Google Calendar (even if work is restricted)\n• Connect Apple Calendar\n• Or skip - I can work with just Granola\n```\n\nUse the same AskUserQuestion and processing logic as Flow A (or CLI fallback).\n\n---\n\n## Flow C: Neither Calendar Nor Granola\n\n**The \"Let's Connect Your Tools\" Flow**\n\n```\n? **Welcome back, [Name]!**\n\nYour workspace is set up, but you don't have calendar or meeting tools connected yet.\n\n**Let's get you integrated with the tools you actually use.**\n\nWhat tools do you use most? Examples:\n• Notion, Linear, Jira\n• Slack, Discord\n• GitHub, GitLab\n• Your company's internal tools\n• Newsletters, RSS feeds\n\n**Give me names or URLs** (URLs are better - just the root domain like \"notion.so\" or \"linear.app\")\n\nOnce you tell me, I'll:\n1. Check for API documentation\n2. Analyze what's possible\n3. Generate working MCP code\n4. Get you integrated in ~2 minutes\n\nWhat's your most important tool?\n```\n\n### Tool Integration Flow\n\nWhen user provides tool name\u002FURL:\n\n**Step 1: Find API docs**\n```python\nif url_provided:\n    # Fetch directly\n    base_url = normalize_url(user_input)\n    doc_content = web_fetch(base_url)\nelse:\n    # Search for docs\n    search_patterns = [\n        f\"{tool_name} API documentation\",\n        f\"{tool_name} developer docs\",\n        f\"{tool_name} API reference\"\n    ]\n    # Try WebSearch or direct fetch of common patterns\n    doc_url = find_api_docs(tool_name)\n    doc_content = web_fetch(doc_url)\n```\n\n**Step 2: Analyze capabilities**\n```\n\"Hold on - reading [Tool]'s API documentation...\"\n\n[Analyze doc_content for:]\n- Authentication methods\n- Available endpoints\n- Common operations\n- Rate limits\n\n\"Got it! I just read [Tool]'s API. Here's what we can build:\n\n**Possible integrations:**\n• [Capability 1] - [Amp use case]\n• [Capability 2] - [Amp use case]\n• [Capability 3] - [Amp use case]\n\nBased on your role ([role]), I'd focus on [specific capabilities].\n\nReady to build this? Takes about 2 minutes.\"\n```\n\n**Step 3: Generate MCP**\n\nFollow `\u002Fcreate-mcp` skill flow:\n1. Design tools based on API analysis\n2. Generate server code\n3. Configure authentication\n4. Test connection\n5. Update documentation\n\n**Step 4: The Magic Moment**\n```\n\"Your [Tool] integration is live! ✅\n\nTest it right now:\n'[Natural language query about that tool]'\n\nSee? Real data, not AI guessing. That's what MCP does.\"\n```\n\n**Step 5: Offer More**\n```\n\"Want to connect another tool? Or explore what else Amp can do?\"\n\n[If yes, repeat]\n[If no, show quick reference guide]\n```\n\n---\n\n## Additional Pathways (Available in Any Flow)\n\n### Google Workspace Setup\n\nIf user wants to connect Gmail\u002FCalendar:\n\n```\n\"Let's connect your Google account.\n\nThis works even if your work account has restrictions - we can use your personal Google if needed.\n\n**What we'll set up:**\n• Calendar sync (see meetings in Amp)\n• Gmail access (summarize newsletters, find emails)\n• Optional: Create daily digest\n\nI'll guide you through OAuth step-by-step.\n\nReady?\"\n```\n\nThen:\n1. Guide through OAuth flow for Google Calendar MCP\n2. Add Gmail MCP if they want\n3. Offer to set up newsletter digest\n\n### Information Diet Setup\n\nIf user mentions newsletters, RSS, or content consumption:\n\n```\n\"Want me to create a daily digest for you?\n\nI can pull from:\n• Newsletters (via Gmail)\n• YouTube channels you follow\n• RSS feeds from blogs\n• Specific websites you check regularly\n\nEach morning, you'd get:\n• Summaries of new content\n• Key points extracted\n• Novel insights highlighted\n• All in one place\n\nWhat sources matter most to you?\"\n```\n\nThen build appropriate scrapers\u002Fintegrations.\n\n### Smithery.ai Discovery\n\nAnytime during tool discussion:\n\n```\n\"By the way - there are 100+ pre-built MCP servers at **Smithery.ai**\n\nBrowse there for:\n• GitHub, Linear, Jira\n• Notion, Airtable, Sheets\n• Slack, Discord, Email\n• Databases, monitoring tools\n• And way more...\n\nFind something interesting? Just paste the URL here and I'll integrate it.\n\nOr we can build custom if you don't find what you need.\"\n```\n\n---\n\n## Setting Expectations\n\nThroughout ANY flow, include:\n\n```\n\"**Fair warning:** I can't promise everything will work perfectly.\n\nCompany restrictions, API limitations, weird auth flows - \nlots can go wrong in the real world.\n\n**But here's what's cool:** Even when it doesn't work, you learn:\n• How to use Claude to debug APIs\n• How to build integrations yourself\n• How MCP servers work\n• How to read API documentation\n\nIt's educational even when it fails.\n\nStill want to try?\"\n```\n\n---\n\n## Cursor UX Tip (If Running in Cursor)\n\nAfter a few file edits, offer:\n\n```\n\"**Quick Cursor tip:**\n\nYou're seeing 'Accept' prompts for each change I make. That's Cursor asking permission.\n\n**Useful at first** - you see what's happening\n**Annoying later** - slows things down\n\nIf you want to auto-accept:\n1. Settings (Cmd+,)\n2. Search \"always allow tool use\"\n3. Enable it\n\nYour choice - some prefer control, others prefer speed.\"\n```\n\nOnly show if:\n- Detected running in Cursor (check environment)\n- After 3-4 accept prompts shown\n\n---\n\n## Integration Discovery (Contextual)\n\n**Trigger:** Run this check if the vault is \u003C 7 days old AND few or no integrations are connected.\n\n### When to Check\n\nAfter completing the main pathway flow (A, B, or C) but before showing the completion message:\n\n1. Read `System\u002Fintegrations\u002Fconfig.yaml` (if it exists)\n2. Count how many integrations have `enabled: true` (exclude `slack` since it's a default)\n3. If connected integrations \u003C= 1 AND vault age \u003C 7 days, run the concierge\n\n### How to Run\n\nSearch the vault for references to tools the user hasn't connected yet (Slack, Todoist, Google Calendar, Zoom, etc.). Check file contents in `00-Inbox\u002F`, `04-Projects\u002F`, and `05-Areas\u002F` for mentions.\n\n### Presenting Recommendations\n\n**Only surface high-value items** (keep it light, this isn't onboarding, it's a nudge).\n\nFor each high_value integration found:\n\n```\nBy the way, I noticed you mention **[shortName]** in a few places\n([mentions] references across files like [first example file]).\n\nConnecting it would give you: [value proposition]\n\nSetup takes [setupTime]. Want to connect it now?\n```\n\n**If multiple high_value items:**\n\n```\nI also spotted signals for **[shortName2]** and **[shortName3]** in your notes.\nWant to connect any of these? Or run `\u002Fintegrate-mcp` anytime later.\n```\n\n**Rules:**\n- Maximum 2 integration suggestions (don't overwhelm)\n- Only mention high_value items (score >= 5) — skip moderate and available\n- If user says yes, run the setup skill inline then return to the getting-started completion\n- If user says no\u002Fskip\u002Flater, move on without pressure\n- Don't show this section if the user already went through integration setup during onboarding Step 8\n\n### Detecting Prior Setup\n\n\n---\n\n## Completion & Next Steps\n\nAfter any pathway completes:\n\n```\n\"**You're set up!**\n\n**Daily workflow:**\n• Run `\u002Fdaily-plan` each morning\n• Use `\u002Fmeeting-prep [person]` before meetings\n• Tell me about meetings - I'll extract action items\n\n**Discovery:**\n• `\u002Famp-level-up` - Find features you haven't tried\n• `\u002Fintegrate-mcp` - Add more tools anytime\n• Smithery.ai - Browse MCP marketplace\n\n**Come back anytime:**\n• `\u002Fgetting-started` - Run this tour again\n• Just ask in natural language - I'll figure out what you need\n\nWhat would you like to work on first?\"\n```\n\nUpdate marker file:\n```python\nmarker_data['phase2_completed'] = True\nmarker_data['phase2_completed_at'] = datetime.now().isoformat()\nmarker_data['pathways_completed'] = selected_pathways\n```\n\n---\n\n## Usage Log Integration\n\nBefore showing suggestions, check `System\u002Fusage_log.md`:\n\n```python\nusage = read_usage_log()\ntried_features = [f for f, checked in usage.items() if checked]\n\n# Adaptive suggestions based on what they've used\nif len(tried_features) \u003C 3:\n    # New user - basics\n    suggest = [\"daily-plan\", \"meeting-prep\"]\nelif \"meeting-prep\" not in tried_features and has_calendar:\n    # Have meetings but haven't prepped\n    suggest = [\"meeting-prep\"]\nelif \"process-meetings\" not in tried_features and has_granola:\n    # Have Granola but haven't processed\n    suggest = [\"process-meetings\"]\nelse:\n    # Power user - advanced features\n    suggest = get_advanced_features(usage, role)\n```\n\nDon't show features they've already tried - focus on gaps.\n\n---\n\n## Escape Hatches\n\nAt EVERY major decision point:\n\n- \"This is a lot - want to pause and come back later?\"\n- \"No pressure - you can always run `\u002Fgetting-started` again\"\n- \"Want to explore on your own first? That's cool\"\n\nAfter full tour:\n- \"Remember - invoke `\u002Famp-level-up` to discover more\"\n- \"Or `\u002Fintegrate-mcp` to add tools as you need them\"\n- \"The system grows with you\"\n\n---\n\n## Success Metric\n\nThe moment they think: **\"How did it know to do that?\"**\n\nDifferent magic for different situations:\n- **With data:** Analyzed calendar\u002FGranola intelligently, offered smart actions\n- **Without data:** Built working integration in 2 minutes from just a name\n- **Either way:** They see value immediately and know where to go next\n\n---\n\n## Helper: Analyze Granola Data Extent\n\nBefore presenting choices, discover how much Granola data is available:\n\n```python\ndef analyze_granola_extent(user_email_domain: str, extended: bool = False) -> dict:\n    \"\"\"\n    Discover how much Granola data is available\n    \n    Args:\n        user_email_domain: User's company email domain for internal\u002Fexternal classification\n        extended: If True, fetch up to 2 years. If False, fetch 6 months (default)\n    \n    Returns:\n        Dictionary with extent analysis or None if no data\n    \"\"\"\n    from datetime import datetime\n    \n    # Default to 6 months for speed, optionally extend to 2 years\n    days_to_fetch = 365 * 2 if extended else 180\n    result = granola_get_recent_meetings(days_back=days_to_fetch, limit=1000)\n    \n    if not result.get('success') or not result.get('meetings'):\n        return {\n            'meetings_count': 0,\n            'days_back': 0,\n            'has_data': False\n        }\n    \n    meetings = result['meetings']\n    \n    # Find oldest and newest dates\n    dates = [m['date'] for m in meetings if m.get('date')]\n    if not dates:\n        return {'meetings_count': 0, 'days_back': 0, 'has_data': False}\n    \n    oldest = min(dates)\n    newest = max(dates)\n    \n    # Calculate days back\n    oldest_dt = datetime.fromisoformat(oldest)\n    newest_dt = datetime.fromisoformat(newest)\n    days_back = (newest_dt - oldest_dt).days + 1  # +1 to include both days\n    \n    # Extract unique people and companies\n    people = set()\n    internal_people = set()\n    external_people = set()\n    companies = set()\n    \n    # Normalize user domain for comparison\n    user_domains = [d.strip().lower() for d in user_email_domain.split(',')]\n    \n    for meeting in meetings:\n        for participant in meeting.get('participants', []):\n            name = participant.get('name')\n            email = participant.get('email')\n            \n            if name:\n                people.add(name)\n                \n                if email:\n                    domain = email.split('@')[1].lower() if '@' in email else None\n                    \n                    # Classify as internal or external\n                    if domain and any(d in domain or domain in d for d in user_domains):\n                        internal_people.add(name)\n                    else:\n                        external_people.add(name)\n                        if domain:\n                            companies.add(domain)\n                else:\n                    # No email provided - default to external\n                    external_people.add(name)\n    \n    # Calculate meetings in different time ranges for estimation\n    now = datetime.now()\n    meetings_7d = sum(1 for m in meetings if m.get('date') and \n                      (now - datetime.fromisoformat(m['date'])).days \u003C= 7)\n    meetings_30d = sum(1 for m in meetings if m.get('date') and \n                       (now - datetime.fromisoformat(m['date'])).days \u003C= 30)\n    meetings_90d = sum(1 for m in meetings if m.get('date') and \n                       (now - datetime.fromisoformat(m['date'])).days \u003C= 90)\n    \n    # Check if there might be more data beyond what we fetched\n    has_more = False\n    if not extended and len(meetings) >= 900:  # Close to limit, likely more data\n        has_more = True\n    \n    # Check if oldest meeting is exactly at the boundary (suggests more data)\n    if not extended and days_back >= 175:  # Close to 180 days\n        has_more = True\n    \n    return {\n        'has_data': True,\n        'meetings_count': len(meetings),\n        'days_back': days_back,\n        'oldest_date': oldest,\n        'newest_date': newest,\n        'unique_people': len(people),\n        'internal_people': len(internal_people),\n        'external_people': len(external_people),\n        'unique_companies': len(companies),\n        'people_sample': list(people)[:10],\n        'companies_list': list(companies),\n        'meetings_7d': meetings_7d,\n        'meetings_30d': meetings_30d,\n        'meetings_90d': meetings_90d,\n        'has_more_data': has_more,\n        'fetched_range_days': days_to_fetch\n    }\n```\n\n**Usage in Flow A:**\n```python\n# After detecting Granola is available\nuser_profile = read_user_profile()\nemail_domain = user_profile.get('email_domain', '')\n\n# Fetch initial 6 months\nextent = analyze_granola_extent(email_domain)\n\nif not extent['has_data']:\n    # Handle no data case\n    pass\nelse:\n    # Check if there's more data beyond 6 months\n    if extent['has_more_data']:\n        # Ask if they want to fetch more\n        response = AskUserQuestion({\n            \"questions\": [{\n                \"id\": \"fetch_more\",\n                \"prompt\": f\"I found {extent['meetings_count']} meetings going back {extent['days_back']} days. There appears to be more data beyond that. Want me to check how much more?\",\n                \"allow_multiple\": false,\n                \"options\": [\n                    {\"id\": \"yes\", \"label\": \"Yes - show me the full extent\"},\n                    {\"id\": \"no\", \"label\": \"No - 6 months is enough\"}\n                ]\n            }]\n        })\n        \n        if response['fetch_more'] == 'yes':\n            # Fetch extended range (2 years)\n            extent = analyze_granola_extent(email_domain, extended=True)\n    \n    # Show discovery summary and choice framework\n    pass\n```\n\n---\n\n## Track Usage (Silent)\n\nUpdate `System\u002Fusage_log.md` to mark getting started as used.\n\n**Analytics (Silent):**\n\nCall `track_event` with event_name `getting_started_completed` and properties:\n- calendar_connected\n- granola_connected\n- artifacts_created\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\u002Fgetting-started",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 getting-started",{"tool":43,"command":44},"bunx","bunx skills add hemory\u002Famp-legacy-do-not-use --skill getting-started",{"tool":46,"command":47},"pnpm","pnpm dlx skills add hemory\u002Famp-legacy-do-not-use --skill getting-started","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp-legacy-do-not-use\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fgetting-started\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"]