{"id":"03035007-43fe-498c-86aa-adfbfb4e167f","shortId":"u9MdfR","kind":"skill","title":"twitter-cultivate","tagline":"Twitter/X account cultivation and growth system. Checks account health (TweepCred, shadowban), analyzes tweets, finds engagement opportunities, recommends unfollows, and tracks progress. Triggers on \"/twitter-cultivate\", \"check my twitter\", \"twitter health\", \"grow my twitter\", \"t","description":"# Twitter Account Cultivation Skill\n\nSystematic approach to growing Twitter presence based on the **open-source algorithm analysis**.\n\n---\n\n## Prerequisites\n\n- **Twikit** installed (`pip install twikit`)\n- **Twitter cookies** exported to: `~/crawlee-social-scraper/twitter_cookies.json`\n  (or wherever your Twikit setup lives — set `TWIKIT_DIR` env var)\n- Your Twitter handle configured in state file\n\n---\n\n## Core Metrics to Track\n\n| Metric | Healthy Range | Impact |\n|--------|---------------|--------|\n| Following/Follower Ratio | **< 0.6** | TweepCred score |\n| Avg Views/Tweet | 20-40% of followers | Algorithm favor |\n| Media Tweet % | **> 50%** | 10x engagement |\n| Link Tweet % | **< 20%** | Avoid algorithm penalty |\n| Reply Rate | Reply to 100% of comments | +75 weight boost |\n\n---\n\n## Workflow: Full Health Check\n\n### Step 1: Analyze Account\n\n```bash\ncd ~/crawlee-social-scraper\nsource venv/bin/activate\npython twitter_cultivate.py analyze <YOUR_USERNAME>\n```\n\n### Step 2: Check Shadowban Status\n\n```bash\npython twitter_cultivate.py shadowban <YOUR_USERNAME>\n```\n\nOr manually check: [shadowban.yuzurisa.com](https://shadowban.yuzurisa.com)\n\n### Step 3: Analyze Following List\n\n```bash\npython twitter_cultivate.py unfollow-recs <YOUR_USERNAME> --limit 50\n```\n\nRecommends accounts to unfollow based on:\n- No tweets in 90+ days (inactive)\n- Never interacted with you (no value)\n- Low follower count + high following (likely bots)\n- No mutual engagement\n\n### Step 4: Find Engagement Opportunities\n\n```bash\npython twitter_cultivate.py opportunities --niche \"AI,startup,founder\"\n```\n\nFinds:\n- Rising tweets in your niche to reply to\n- Accounts similar to yours to engage with\n- Trending topics you can contribute to\n\n### Step 5: Generate Weekly Report\n\n```bash\npython twitter_cultivate.py report\n```\n\n---\n\n## Account Health Scoring\n\nBased on Twitter's open-source algorithm:\n\n### TweepCred Estimation\n\n```\nScore = PageRank × (1 / max(1, following/followers))\n```\n\n| Ratio | Estimated TweepCred | Algorithm Treatment |\n|-------|---------------------|---------------------|\n| < 0.6 | 65+ (healthy) | All tweets considered |\n| 0.6 - 2.0 | 40-65 | Limited consideration |\n| 2.0 - 5.0 | 20-40 | Severe penalty |\n| > 5.0 | < 20 | **Only 3 tweets max** |\n\n---\n\n## Unfollow Strategy\n\n### Priority 1: Inactive Accounts\n- No tweets in 90+ days\n- Safe to unfollow, no relationship loss\n\n### Priority 2: Non-Engagers\n- Never liked/replied to your tweets\n- One-way relationship\n\n### Priority 3: Low-Value Follows\n- High following/low followers (bot-like)\n- No content in your niche\n\n### Execution Plan\n\n```\nWeek 1: Unfollow 30 inactive accounts\nWeek 2: Unfollow 30 non-engagers\nWeek 3: Unfollow 30 low-value follows\nWeek 4: Evaluate ratio improvement\n```\n\n**Target:** Get ratio below 2.0, ideally below 0.6\n\n---\n\n## Content Strategy (Algorithm-Optimized)\n\n### Tweet Types by Algorithm Weight\n\n| Type | Weight | Recommendation |\n|------|--------|----------------|\n| Tweet that gets author reply | **+75** | ALWAYS reply to comments |\n| Tweet with replies | +13.5 | Ask questions |\n| Tweet with profile clicks | +12.0 | Be intriguing |\n| Tweet with long dwell time | +10.0 | Use threads |\n| Retweet | +1.0 | Low value |\n| Like | +0.5 | Lowest value |\n\n### Content Mix\n\n- **40%** Value content (insights, tips, frameworks)\n- **30%** Engagement bait (questions, polls, hot takes)\n- **20%** Build-in-public (progress updates, wins, losses)\n- **10%** Promotion (with value attached)\n\n### Media Requirements\n\nEvery tweet should have ONE of:\n- Image (infographic, screenshot, meme)\n- Video (< 2:20, hook in first 3 sec)\n- Poll\n- Thread (7-10 tweets)\n\n**NEVER post text-only tweets**\n\n---\n\n## Posting Schedule\n\n### Optimal Times (General)\n\n| Day | Best Time | Second Best |\n|-----|-----------|-------------|\n| Tuesday | 9-10 AM | 1-2 PM |\n| Wednesday | 9-10 AM | 3-4 PM |\n| Thursday | 10-11 AM | 2-3 PM |\n\n### First 10 Minutes Protocol\n\n```\n1. Post at optimal time\n2. Immediately self-reply with additional insight\n3. Reply to ANY comment within 10 minutes\n4. Have 2-3 \"pod\" members ready to RT\n```\n\n### Frequency\n\n- **Minimum:** 1 tweet/day\n- **Optimal:** 3-5 tweets/day\n- **Gap:** 30-60 min between tweets\n\n---\n\n## Engagement Tactics\n\n### Reply Strategy (Most Important)\n\nThe algorithm gives **+75 weight** when you reply to replies on your tweets.\n\n```\nSomeone comments on your tweet\n    ↓\nReply within 30 minutes (CRITICAL)\n    ↓\nAlgorithm sees author engagement\n    ↓\nTweet gets boosted to more feeds\n```\n\n### Quote Tweet Strategy\n\n```\nFind viral tweet in your niche\n    ↓\nQuote with your unique take\n    ↓\nAdd value, not just \"great point\"\n    ↓\nPost during optimal hours\n```\n\n### Thread Formula\n\n```\n1/ Hook (curiosity gap or bold claim)\n2-6/ Individual points with specifics\n7/ Summary\n8/ CTA: Question or \"follow for more\"\n```\n\n---\n\n## Real-time Monitor & Auto-Reply\n\nMonitor target accounts and auto-generate contextual replies for maximum engagement.\n\n```bash\ncd ~/crawlee-social-scraper\n\n# Interactive mode (recommended) - approve before posting\n./start_monitor.sh\n\n# Monitor only - just watch, no replies\n./start_monitor.sh monitor\n```\n\n### Customize Targets\n\n```bash\n./venv/bin/python3 twitter_monitor.py --interactive --targets \"levelsio,marc_lou,thedankoe\"\n```\n\n### How It Works\n\n1. Polls target accounts every 60 seconds\n2. Detects new tweets\n3. Generates contextual reply\n4. Shows you the suggestion\n5. You approve/edit/skip\n\n---\n\n## CLI Reference\n\n```bash\ncd ~/crawlee-social-scraper\nsource venv/bin/activate\n\n# Full health check\npython twitter_cultivate.py analyze <username>\n\n# Check shadowban\npython twitter_cultivate.py shadowban <username>\n\n# Get unfollow recommendations\npython twitter_cultivate.py unfollow-recs <username> --limit 50\n\n# Find engagement opportunities\npython twitter_cultivate.py opportunities --niche \"AI,startup\"\n\n# Weekly report\npython twitter_cultivate.py report\n\n# Execute unfollow (with confirmation)\npython twitter_cultivate.py unfollow <username> --execute\n```\n\n---\n\n## Weekly Routine\n\n### Daily (15 min)\n\n- [ ] Post 1-3 tweets with media\n- [ ] Reply to ALL comments on your tweets\n- [ ] Engage with 5-10 tweets in your niche\n- [ ] Check notifications and respond\n\n### Weekly (Saturday)\n\n- [ ] Run full health check\n- [ ] Review what content performed best\n- [ ] Unfollow 10-20 low-value accounts\n- [ ] Plan next week's content themes\n\n### Monthly\n\n- [ ] Full ratio review (target < 2.0)\n- [ ] Shadowban check\n- [ ] Content audit (media %, link %)\n- [ ] Milestone check (follower goals)\n\n---\n\n## Recovery Plan (Low Follower Count)\n\n### Phase 1: Emergency Ratio Fix (Week 1-2)\n\nIf your ratio is > 5.0 (following >> followers):\n- Unfollow 100+ inactive/non-engaging accounts\n- Target: ratio < 5.0 as first milestone\n\n### Phase 2: Content Upgrade (Week 2-4)\n\nIf you have 0% media tweets:\n- Add image/video to EVERY tweet\n- Use Canva/Figma for quick graphics\n- Screenshot interesting data/insights\n\n### Phase 3: Engagement Building (Week 3-6)\n\n- Reply to 20+ tweets/day in your niche\n- Quote tweet viral content with your take\n- Join relevant Twitter communities\n- DM potential collaborators\n\n### Phase 4: Consistency (Ongoing)\n\n- 3-5 tweets/day\n- Reply to 100% of comments\n- Weekly analysis and adjustment","tags":["twitter","cultivate","claude","agent","skills","phy041","agent-skills","ai-agents","automation","founder","indie-hacker","openclaw"],"capabilities":["skill","source-phy041","skill-twitter-cultivate","topic-agent-skills","topic-ai-agents","topic-automation","topic-claude","topic-founder","topic-indie-hacker","topic-openclaw","topic-social-media"],"categories":["claude-agent-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/PHY041/claude-agent-skills/twitter-cultivate","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add PHY041/claude-agent-skills","source_repo":"https://github.com/PHY041/claude-agent-skills","install_from":"skills.sh"}},"qualityScore":"0.456","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 13 github stars · SKILL.md body (7,235 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-04-22T13:03:32.728Z","embedding":null,"createdAt":"2026-04-19T00:41:13.577Z","updatedAt":"2026-04-22T13:03:32.728Z","lastSeenAt":"2026-04-22T13:03:32.728Z","tsv":"'+0.5':422 '+1.0':418 '+10.0':414 '+12.0':406 '+13.5':399 '+75':123,391,573 '-10':477,497,504,796 '-11':511 '-2':500,857 '-20':818 '-3':514,544,782 '-4':507,881 '-40':100,280 '-5':556,934 '-6':637,907 '-60':560 '-65':274 '/crawlee-social-scraper':136,672,729 '/crawlee-social-scraper/twitter_cookies.json':65 '/start_monitor.sh':679,686 '/twitter-cultivate':27 '/venv/bin/python3':691 '0':885 '0.6':94,265,271,372 '1':131,256,258,292,340,499,520,552,629,702,781,851,856 '10':449,510,517,539,817 '100':120,866,938 '10x':108 '15':778 '2':143,307,346,467,513,525,543,636,709,876,880 '2.0':272,277,369,834 '20':99,112,279,284,440,468,910 '3':157,286,321,353,472,506,533,555,713,902,906,933 '30':342,348,355,433,559,590 '4':198,361,541,717,930 '40':273,427 '5':233,722,795 '5.0':278,283,862,871 '50':107,168,752 '60':707 '65':266 '7':476,642 '8':644 '9':496,503 '90':178,298 'account':5,11,38,133,170,219,241,294,344,660,705,822,868 'add':617,888 'addit':531 'adjust':944 'ai':207,760 'algorithm':53,103,114,251,263,376,381,571,593 'algorithm-optim':375 'alway':392 'analysi':54,942 'analyz':15,132,141,158,737 'approach':42 'approv':676 'approve/edit/skip':724 'ask':400 'attach':453 'audit':838 'author':389,595 'auto':656,663 'auto-gener':662 'auto-repli':655 'avg':97 'avoid':113 'bait':435 'base':47,173,244 'bash':134,147,161,202,237,670,690,727 'best':491,494,815 'bold':634 'boost':125,599 'bot':193,330 'bot-lik':329 'build':442,904 'build-in-publ':441 'canva/figma':894 'cd':135,671,728 'check':10,28,129,144,153,734,738,801,810,836,842 'claim':635 'cli':725 'click':405 'collabor':928 'comment':122,395,537,584,789,940 'communiti':925 'configur':80 'confirm':770 'consid':270 'consider':276 'consist':931 'content':333,373,425,429,813,827,837,877,918 'contextu':665,715 'contribut':230 'cooki':62 'core':84 'count':189,849 'critic':592 'cta':645 'cultiv':3,6,39 'curios':631 'custom':688 'daili':777 'data/insights':900 'day':179,299,490 'detect':710 'dir':74 'dm':926 'dwell':412 'emerg':852 'engag':18,109,196,200,224,310,351,434,564,596,669,754,793,903 'env':75 'estim':253,261 'evalu':362 'everi':456,706,891 'execut':337,767,774 'export':63 'favor':104 'feed':602 'file':83 'find':17,199,210,606,753 'first':471,516,873 'fix':854 'follow':102,159,188,191,325,328,359,648,843,848,863,864 'following/follower':92 'following/followers':259 'following/low':327 'formula':628 'founder':209 'framework':432 'frequenc':550 'full':127,732,808,830 'gap':558,632 'general':489 'generat':234,664,714 'get':366,388,598,743 'give':572 'goal':844 'graphic':897 'great':621 'grow':33,44 'growth':8 'handl':79 'health':12,32,128,242,733,809 'healthi':89,267 'high':190,326 'hook':469,630 'hot':438 'hour':626 'ideal':370 'imag':462 'image/video':889 'immedi':526 'impact':91 'import':569 'improv':364 'inact':180,293,343 'inactive/non-engaging':867 'individu':638 'infograph':463 'insight':430,532 'instal':57,59 'interact':182,673,693 'interest':899 'intrigu':408 'join':922 'levelsio':695 'like':192,331,421 'liked/replied':312 'limit':167,275,751 'link':110,840 'list':160 'live':71 'long':411 'loss':305,448 'lou':697 'low':187,323,357,419,820,847 'low-valu':322,356,819 'lowest':423 'manual':152 'marc':696 'max':257,288 'maximum':668 'media':105,454,785,839,886 'member':546 'meme':465 'metric':85,88 'mileston':841,874 'min':561,779 'minimum':551 'minut':518,540,591 'mix':426 'mode':674 'monitor':654,658,680,687 'month':829 'mutual':195 'never':181,311,479 'new':711 'next':824 'nich':206,215,336,611,759,800,914 'non':309,350 'non-engag':308,349 'notif':802 'one':317,460 'one-way':316 'ongo':932 'open':51,249 'open-sourc':50,248 'opportun':19,201,205,755,758 'optim':377,487,523,554,625 'pagerank':255 'penalti':115,282 'perform':814 'phase':850,875,901,929 'pip':58 'plan':338,823,846 'pm':501,508,515 'pod':545 'point':622,639 'poll':437,474,703 'post':480,485,521,623,678,780 'potenti':927 'prerequisit':55 'presenc':46 'prioriti':291,306,320 'profil':404 'progress':24,445 'promot':450 'protocol':519 'public':444 'python':139,148,162,203,238,735,740,746,756,764,771 'question':401,436,646 'quick':896 'quot':603,612,915 'rang':90 'rate':117 'ratio':93,260,363,367,831,853,860,870 'readi':547 'real':652 'real-tim':651 'rec':166,750 'recommend':20,169,385,675,745 'recoveri':845 'refer':726 'relationship':304,319 'relev':923 'repli':116,118,217,390,393,398,529,534,566,577,579,588,657,666,685,716,786,908,936 'report':236,240,763,766 'requir':455 'respond':804 'retweet':417 'review':811,832 'rise':211 'routin':776 'rt':549 'run':807 'safe':300 'saturday':806 'schedul':486 'score':96,243,254 'screenshot':464,898 'sec':473 'second':493,708 'see':594 'self':528 'self-repli':527 'set':72 'setup':70 'sever':281 'shadowban':14,145,150,739,742,835 'shadowban.yuzurisa.com':154,155 'show':718 'similar':220 'skill':40 'skill-twitter-cultivate' 'someon':583 'sourc':52,137,250,730 'source-phy041' 'specif':641 'startup':208,761 'state':82 'status':146 'step':130,142,156,197,232 'strategi':290,374,567,605 'suggest':721 'summari':643 'system':9 'systemat':41 'tactic':565 'take':439,616,921 'target':365,659,689,694,704,833,869 'text':482 'text-on':481 'thedanko':698 'theme':828 'thread':416,475,627 'thursday':509 'time':413,488,492,524,653 'tip':431 'topic':227 'topic-agent-skills' 'topic-ai-agents' 'topic-automation' 'topic-claude' 'topic-founder' 'topic-indie-hacker' 'topic-openclaw' 'topic-social-media' 'track':23,87 'treatment':264 'trend':226 'trigger':25 'tuesday':495 'tweepcr':13,95,252,262 'tweet':16,106,111,176,212,269,287,296,315,378,386,396,402,409,457,478,484,563,582,587,597,604,608,712,783,792,797,887,892,916 'tweet/day':553 'tweets/day':557,911,935 'twikit':56,60,69,73 'twitter':2,30,31,35,37,45,61,78,246,924 'twitter-cultiv':1 'twitter/x':4 'twitter_cultivate.py':140,149,163,204,239,736,741,747,757,765,772 'twitter_monitor.py':692 'type':379,383 'unfollow':21,165,172,289,302,341,347,354,744,749,768,773,816,865 'unfollow-rec':164,748 'uniqu':615 'updat':446 'upgrad':878 'use':415,893 'valu':186,324,358,420,424,428,452,618,821 'var':76 'venv/bin/activate':138,731 'video':466 'views/tweet':98 'viral':607,917 'watch':683 'way':318 'wednesday':502 'week':235,339,345,352,360,762,775,805,825,855,879,905,941 'weight':124,382,384,574 'wherev':67 'win':447 'within':538,589 'work':701 'workflow':126","prices":[{"id":"fd58c8bc-aa06-4140-a214-6f48c640b363","listingId":"03035007-43fe-498c-86aa-adfbfb4e167f","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"PHY041","category":"claude-agent-skills","install_from":"skills.sh"},"createdAt":"2026-04-19T00:41:13.577Z"}],"sources":[{"listingId":"03035007-43fe-498c-86aa-adfbfb4e167f","source":"github","sourceId":"PHY041/claude-agent-skills/twitter-cultivate","sourceUrl":"https://github.com/PHY041/claude-agent-skills/tree/main/skills/twitter-cultivate","isPrimary":false,"firstSeenAt":"2026-04-19T00:41:13.577Z","lastSeenAt":"2026-04-22T13:03:32.728Z"}],"details":{"listingId":"03035007-43fe-498c-86aa-adfbfb4e167f","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"PHY041","slug":"twitter-cultivate","github":{"repo":"PHY041/claude-agent-skills","stars":13,"topics":["agent-skills","ai-agents","automation","claude","founder","indie-hacker","openclaw","social-media"],"license":null,"html_url":"https://github.com/PHY041/claude-agent-skills","pushed_at":"2026-02-24T15:25:20Z","description":"Collection of Claude Code Agent Skills for founders, indie hackers, and growth engineers","skill_md_sha":"4a59356e5e65fc48cd5e45feb144becaf0e917f7","skill_md_path":"skills/twitter-cultivate/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/PHY041/claude-agent-skills/tree/main/skills/twitter-cultivate"},"layout":"multi","source":"github","category":"claude-agent-skills","frontmatter":{"name":"twitter-cultivate","description":"Twitter/X account cultivation and growth system. Checks account health (TweepCred, shadowban), analyzes tweets, finds engagement opportunities, recommends unfollows, and tracks progress. Triggers on \"/twitter-cultivate\", \"check my twitter\", \"twitter health\", \"grow my twitter\", \"twitter maintenance\", \"fix my twitter reach\"."},"skills_sh_url":"https://skills.sh/PHY041/claude-agent-skills/twitter-cultivate"},"updatedAt":"2026-04-22T13:03:32.728Z"}}