{"id":"871808bf-8817-4345-9e54-d4c827d85d6f","shortId":"37LFba","kind":"skill","title":"product-taste-intuition","tagline":"Build product taste via a Taste Calibration Sprint (benchmarks, critique notes, hypothesis log).","description":"# Product Taste & Intuition\n\n## Scope\n\n**Covers**\n- Developing **product taste** (what \"good\" looks like) through deliberate exposure, observation, and critique\n- Using intuition as a **hypothesis generator** (turning \"gut feel\" into testable hypotheses)\n- Building a repeatable **practice loop** (exposure hours → analysis → validation → updated taste rules)\n\n**When to use**\n- \"Help me improve my product taste / product sense.\"\n- \"Calibrate what ‘good onboarding’ looks like for our product category.\"\n- \"Turn my intuition about this flow into testable hypotheses.\"\n- \"Create a structured way to study great products and extract patterns.\"\n\n**When NOT to use**\n- You need to decide *what to build* (use `problem-definition`, `prioritizing-roadmap`, or `defining-product-vision`).\n- You need user evidence first (use `conducting-user-interviews` or `usability-testing`).\n- You want aesthetic critique only (this is product experience: value, UX, clarity, trust, speed--not just visuals).\n- You can’t name any target user, use case, or the \"taste domain\" you want to improve (we’ll narrow first).\n- You want to run a structured design review of your own product’s UI/UX (use `running-design-reviews`).\n- You want to dogfood your own product and capture feedback (use `dogfooding`).\n- You want to develop a PM’s skills through coaching conversations (use `coaching-pms`).\n- You need a competitive landscape analysis or market positioning (use `competitive-analysis`).\n\n## Inputs\n\n**Minimum required**\n- Taste domain to improve (pick 1): onboarding, activation, navigation/IA, editor/workflow, pricing/packaging UX, notifications, retention loops, trust/safety, performance/latency feel, copy/voice\n- Target user + top job-to-be-done for that domain\n- 3–10 benchmark products/experiences to study (or \"unknown—please propose\")\n- Time box (e.g., 60–120 min sprint; or a 2–4 week practice plan)\n- Constraints (platform, geography, accessibility, compliance, brand voice, etc.)\n\n**Missing-info strategy**\n- Ask up to 5 questions from [references/INTAKE.md](references/INTAKE.md).\n- If inputs remain missing, proceed with explicit assumptions and provide 2 scope options (narrow vs broad).\n\n## Outputs (deliverables)\n\nProduce a **Taste Calibration Pack** (in-chat Markdown; or as files if requested):\n\n1) **Taste Calibration Brief** (domain, target user/job, what \"good\" means, constraints)\n2) **Benchmark Set** (5–10 products) + \"why these\" + what to study\n3) **Product Study Notes** (1 page per benchmark) using a consistent critique template\n4) **Taste Rules + Anti-Patterns** (do/don’t rules derived from evidence)\n5) **Intuition → Hypothesis Log** (testable hypotheses + predicted signals)\n6) **Validation Plan** (qual + quant checks; smallest viable tests)\n7) **Practice Plan** (2–4 weeks: exposure hours + weekly synthesis cadence)\n8) **Risks / Open questions / Next steps** (always included)\n\nTemplates: [references/TEMPLATES.md](references/TEMPLATES.md)\n\n## Workflow (8 steps)\n\n### 1) Intake + pick the taste domain (narrow the problem)\n- **Inputs:** User context; [references/INTAKE.md](references/INTAKE.md).\n- **Actions:** Choose 1 taste domain and 1 \"moment\" (e.g., first-run onboarding). Define target user + job + constraints. Set time box.\n- **Outputs:** Taste Calibration Brief (draft).\n- **Checks:** A stakeholder can answer: \"What specific experience are we calibrating taste for?\"\n\n### 2) Define \"good taste\" as decision criteria (not vibes)\n- **Inputs:** Domain + user/job.\n- **Actions:** Draft 6–10 criteria (e.g., clarity, time-to-value, trust, agency, error recovery, perceived speed, cognitive load). Add explicit tradeoffs (what you’ll sacrifice).\n- **Outputs:** Criteria list + tradeoffs section in the brief.\n- **Checks:** Criteria are observable in-product (you can point to UI/behavior), not generic adjectives.\n\n### 3) Build the benchmark set (exposure hours, curated)\n- **Inputs:** Known benchmarks (or none).\n- **Actions:** Select 5–10 exemplars (direct, adjacent, and at least 1 \"gold standard\"). For each: what you’re studying and why it’s relevant.\n- **Outputs:** Benchmark Set table.\n- **Checks:** Set includes at least 2 \"outside the category\" references to avoid local maxima.\n\n### 4) Study like a voracious user (structured observation)\n- **Inputs:** Benchmarks; critique template.\n- **Actions:** Use each product as the target user. Capture micro-moments: friction, delight, confusion, trust breaks. Record \"what happened\" before \"why it’s good/bad\".\n- **Outputs:** Product Study Notes (draft).\n- **Checks:** Each benchmark note includes at least 3 concrete moments with screenshots/quotes if available (or precise descriptions).\n\n### 5) Synthesize: turn observations into taste rules + anti-patterns\n- **Inputs:** Study notes across benchmarks.\n- **Actions:** Cluster patterns. Convert into rules: **DO/DO NOT**, plus rationale and where it applies. Add anti-patterns that create \"AI slop\" (generic, incoherent, misaligned experiences).\n- **Outputs:** Taste Rules + Anti-Patterns.\n- **Checks:** Each rule is backed by ≥ 2 observations from different benchmarks (or explicitly marked \"hypothesis\").\n\n### 6) Intuition as hypothesis generator (make it testable)\n- **Inputs:** Rules + your gut reactions.\n- **Actions:** Write intuition statements (\"It feels off because…\") and convert into testable hypotheses with predicted signals and counter-signals.\n- **Outputs:** Intuition → Hypothesis Log.\n- **Checks:** Each hypothesis has a clear falsification condition (\"If X doesn’t change after Y, we were wrong.\").\n\n### 7) Validate with smallest viable checks (qual + quant)\n- **Inputs:** Hypothesis log; available data/research access.\n- **Actions:** Choose the lightest validation per hypothesis: usability task, intercept prompt, session replay review, funnel slice, A/B smoke test, copy test, etc. Define success metrics and sample.\n- **Outputs:** Validation Plan with owners/cadence if known.\n- **Checks:** Validation steps are feasible within the stated time box and don’t require sensitive data.\n\n### 8) Create a practice loop + quality gate + finalize\n- **Inputs:** Draft pack.\n- **Actions:** Build a 2–4 week practice plan (exposure hours schedule + weekly synthesis). Run [references/CHECKLISTS.md](references/CHECKLISTS.md) and score with [references/RUBRIC.md](references/RUBRIC.md). Add Risks/Open questions/Next steps.\n- **Outputs:** Final Taste Calibration Pack.\n- **Checks:** A reader can follow the practice plan without additional context; assumptions are explicit.\n\n## Quality gate (required)\n- Use [references/CHECKLISTS.md](references/CHECKLISTS.md) and [references/RUBRIC.md](references/RUBRIC.md).\n- Always include: **Risks**, **Open questions**, **Next steps**.\n\n## Examples\n\n**Example 1 (Onboarding):** \"Calibrate our onboarding taste vs best-in-class. Target users are first-time PMs. Time box: 90 minutes. Output a Taste Calibration Pack.\"  \nExpected: benchmark set, critique notes, taste rules, hypotheses, and a lightweight validation plan.\n\n**Example 2 (B2B workflow UX):** \"My gut says our ‘create project’ flow feels slow and confusing. Turn that into testable hypotheses and a validation plan.\"  \nExpected: intuition→hypothesis log with falsification conditions and smallest viable checks.\n\n**Boundary example:** \"Tell me what good taste is in general.\"\nResponse: require a specific domain + target user/job; otherwise produce a menu of domain options and propose a narrow starting point.\n\n**Boundary example 2:** \"Review our product's dashboard design and tell me what's wrong.\"\nResponse: critiquing your own product's specific design is a design review, not taste calibration. Use `running-design-reviews` for structured UI/UX critique, or `dogfooding` to capture user-perspective feedback.\n\n## Anti-patterns (common failure modes)\n\n1. **Benchmark tourism**: Skimming 10 products in 20 minutes without structured observation. Taste calibration requires deep, moment-by-moment study, not surface-level impressions.\n2. **Opinion without observation**: Writing critique notes that say \"this feels good\" without recording specific micro-moments (friction, delight, confusion, trust breaks). Observations come before interpretations.\n3. **Local maxima benchmarking**: Only studying products in your exact category. Including 2+ \"outside the category\" references prevents copying competitors and reveals transferable patterns.\n4. **Unfalsifiable taste rules**: Deriving rules like \"the UX should be intuitive\" with no way to test or disprove them. Every taste rule should convert to a testable hypothesis with a clear falsification condition.\n5. **One-off sprint with no practice loop**: Running a single calibration session and calling it done. Taste is built through repeated exposure; the practice plan must include a weekly synthesis cadence over 2-4 weeks.","tags":["product","taste","intuition","lenny","skills","plus","liqiongyu","agent-skills","ai-agents","automation","claude","codex"],"capabilities":["skill","source-liqiongyu","skill-product-taste-intuition","topic-agent-skills","topic-ai-agents","topic-automation","topic-claude","topic-codex","topic-prompt-engineering","topic-refoundai","topic-skillpack"],"categories":["lenny_skills_plus"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/liqiongyu/lenny_skills_plus/product-taste-intuition","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add liqiongyu/lenny_skills_plus","source_repo":"https://github.com/liqiongyu/lenny_skills_plus","install_from":"skills.sh"}},"qualityScore":"0.474","qualityRationale":"deterministic score 0.47 from registry signals: · indexed on github topic:agent-skills · 49 github stars · SKILL.md body (8,820 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":"2026-04-22T00:56:24.655Z","embedding":null,"createdAt":"2026-04-18T22:16:58.010Z","updatedAt":"2026-04-22T00:56:24.655Z","lastSeenAt":"2026-04-22T00:56:24.655Z","tsv":"'-4':1220 '1':243,344,370,433,449,453,570,915,1074 '10':269,359,501,563,1078 '120':282 '2':287,322,355,411,486,593,714,856,956,1023,1100,1139,1219 '20':1081 '3':268,366,547,651,1127 '4':288,379,412,602,857,1151 '5':307,358,391,562,661,1185 '6':399,500,723 '60':281 '7':408,778 '8':419,431,842 '90':935 'a/b':808 'access':295,791 'across':674 'action':447,498,560,614,676,736,792,853 'activ':245 'add':517,690,874 'addit':892 'adjac':566 'adject':546 'aesthet':140 'agenc':510 'ai':696 'alway':425,906 'analysi':55,227,234 'answer':477 'anti':383,669,692,706,1069 'anti-pattern':382,668,691,705,1068 'appli':689 'ask':304 'assumpt':319,894 'avail':657,789 'avoid':599 'b2b':957 'back':712 'benchmark':13,270,356,373,550,557,585,611,646,675,718,943,1075,1130 'best':923 'best-in-class':922 'boundari':991,1021 'box':279,467,835,934 'brand':297 'break':630,1122 'brief':347,471,531 'broad':327 'build':5,48,111,548,854 'built':1205 'cadenc':418,1217 'calibr':11,71,333,346,470,483,881,917,940,1050,1087,1197 'call':1200 'captur':203,622,1063 'case':163 'categori':80,596,1137,1142 'chang':772 'chat':337 'check':404,473,532,588,644,708,760,783,826,883,990 'choos':448,793 'clariti':149,504 'class':925 'clear':765,1182 'cluster':677 'coach':216,220 'coaching-pm':219 'cognit':515 'come':1124 'common':1071 'competit':225,233 'competitive-analysi':232 'competitor':1146 'complianc':296 'concret':652 'condit':767,986,1184 'conduct':131 'conducting-user-interview':130 'confus':628,970,1120 'consist':376 'constraint':292,354,464 'context':444,893 'convers':217 'convert':679,745,1175 'copi':811,1145 'copy/voice':256 'counter':754 'counter-sign':753 'cover':22 'creat':90,695,843,964 'criteria':492,502,525,533 'critiqu':14,35,141,377,612,945,1037,1059,1105 'curat':554 'dashboard':1028 'data':841 'data/research':790 'decid':108 'decis':491 'deep':1089 'defin':121,460,487,814 'defining-product-vis':120 'definit':115 'deliber':31 'delight':627,1119 'deliver':329 'deriv':388,1155 'descript':660 'design':182,193,1029,1043,1046,1054 'develop':23,210 'differ':717 'direct':565 'disprov':1169 'do/do':682 'do/don':385 'doesn':770 'dogfood':198,206,1061 'domain':167,239,267,348,438,451,496,1005,1013 'done':264,1202 'draft':472,499,643,851 'e.g':280,455,503 'editor/workflow':247 'error':511 'etc':299,813 'everi':1171 'evid':127,390 'exact':1136 'exampl':913,914,955,992,1022 'exemplar':564 'expect':942,980 'experi':146,480,701 'explicit':318,518,720,896 'exposur':32,53,414,552,861,1208 'extract':99 'failur':1072 'falsif':766,985,1183 'feasibl':830 'feedback':204,1067 'feel':44,255,741,967,1110 'file':341 'final':849,879 'first':128,175,457,930 'first-run':456 'first-tim':929 'flow':86,966 'follow':887 'friction':626,1118 'funnel':806 'gate':848,898 'general':1000 'generat':41,727 'generic':545,698 'geographi':294 'gold':571 'good':27,73,352,488,996,1111 'good/bad':638 'great':96 'gut':43,734,961 'happen':633 'help':63 'hour':54,415,553,862 'hypothes':47,89,396,748,949,975 'hypothesi':16,40,393,722,726,758,762,787,798,982,1179 'impress':1099 'improv':65,171,241 'in-chat':335 'in-product':536 'includ':426,590,648,907,1138,1213 'incoher':699 'info':302 'input':235,313,442,495,555,610,671,731,786,850 'intak':434 'intercept':801 'interpret':1126 'interview':133 'intuit':4,20,37,83,392,724,738,757,981,1162 'job':261,463 'job-to-be-don':260 'known':556,825 'landscap':226 'least':569,592,650 'level':1098 'lightest':795 'lightweight':952 'like':29,76,604,1157 'list':526 'll':173,522 'load':516 'local':600,1128 'log':17,394,759,788,983 'look':28,75 'loop':52,252,846,1193 'make':728 'mark':721 'markdown':338 'market':229 'maxima':601,1129 'mean':353 'menu':1011 'metric':816 'micro':624,1116 'micro-mo':623,1115 'min':283 'minimum':236 'minut':936,1082 'misalign':700 'miss':301,315 'missing-info':300 'mode':1073 'moment':454,625,653,1091,1093,1117 'moment-by-mo':1090 'must':1212 'name':158 'narrow':174,325,439,1018 'navigation/ia':246 'need':106,125,223 'next':423,911 'none':559 'note':15,369,642,647,673,946,1106 'notif':250 'observ':33,535,609,664,715,1085,1103,1123 'onboard':74,244,459,916,919 'one':1187 'one-off':1186 'open':421,909 'opinion':1101 'option':324,1014 'otherwis':1008 'output':328,468,524,584,639,702,756,819,878,937 'outsid':594,1140 'owners/cadence':823 'pack':334,852,882,941 'page':371 'pattern':100,384,670,678,693,707,1070,1150 'per':372,797 'perceiv':513 'performance/latency':254 'perspect':1066 'pick':242,435 'plan':291,401,410,821,860,890,954,979,1211 'platform':293 'pleas':276 'plus':684 'pm':212 'pms':221,932 'point':541,1020 'posit':230 'practic':51,290,409,845,859,889,1192,1210 'precis':659 'predict':397,750 'prevent':1144 'pricing/packaging':248 'priorit':117 'prioritizing-roadmap':116 'problem':114,441 'problem-definit':113 'proceed':316 'produc':330,1009 'product':2,6,18,24,67,69,79,97,122,145,187,201,360,367,538,617,640,1026,1040,1079,1133 'product-taste-intuit':1 'products/experiences':271 'project':965 'prompt':802 'propos':277,1016 'provid':321 'qual':402,784 'qualiti':847,897 'quant':403,785 'question':308,422,910 'questions/next':876 'rational':685 're':577 'reaction':735 'reader':885 'record':631,1113 'recoveri':512 'refer':597,1143 'references/checklists.md':867,868,901,902 'references/intake.md':310,311,445,446 'references/rubric.md':872,873,904,905 'references/templates.md':428,429 'relev':583 'remain':314 'repeat':50,1207 'replay':804 'request':343 'requir':237,839,899,1002,1088 'respons':1001,1036 'retent':251 'reveal':1148 'review':183,194,805,1024,1047,1055 'risk':420,908 'risks/open':875 'roadmap':118 'rule':59,381,387,667,681,704,710,732,948,1154,1156,1173 'run':179,192,458,866,1053,1194 'running-design-review':191,1052 'sacrific':523 'sampl':818 'say':962,1108 'schedul':863 'scope':21,323 'score':870 'screenshots/quotes':655 'section':528 'select':561 'sens':70 'sensit':840 'session':803,1198 'set':357,465,551,586,589,944 'signal':398,751,755 'singl':1196 'skill':214 'skill-product-taste-intuition' 'skim':1077 'slice':807 'slop':697 'slow':968 'smallest':405,781,988 'smoke':809 'source-liqiongyu' 'specif':479,1004,1042,1114 'speed':151,514 'sprint':12,284,1189 'stakehold':475 'standard':572 'start':1019 'state':833 'statement':739 'step':424,432,828,877,912 'strategi':303 'structur':92,181,608,1057,1084 'studi':95,273,365,368,578,603,641,672,1094,1132 'success':815 'surfac':1097 'surface-level':1096 'synthes':662 'synthesi':417,865,1216 'tabl':587 'target':160,257,349,461,620,926,1006 'task':800 'tast':3,7,10,19,25,58,68,166,238,332,345,380,437,450,469,484,489,666,703,880,920,939,947,997,1049,1086,1153,1172,1203 'tell':993,1031 'templat':378,427,613 'test':137,407,810,812,1167 'testabl':46,88,395,730,747,974,1178 'time':278,466,506,834,931,933 'time-to-valu':505 'top':259 'topic-agent-skills' 'topic-ai-agents' 'topic-automation' 'topic-claude' 'topic-codex' 'topic-prompt-engineering' 'topic-refoundai' 'topic-skillpack' 'tourism':1076 'tradeoff':519,527 'transfer':1149 'trust':150,509,629,1121 'trust/safety':253 'turn':42,81,663,971 'ui/behavior':543 'ui/ux':189,1058 'unfalsifi':1152 'unknown':275 'updat':57 'usability-test':135 'usabl':136,799 'use':36,62,104,112,129,162,190,205,218,231,374,615,900,1051 'user':126,132,161,258,443,462,607,621,927,1065 'user-perspect':1064 'user/job':350,497,1007 'ux':148,249,959,1159 'valid':56,400,779,796,820,827,953,978 'valu':147,508 'via':8 'viabl':406,782,989 'vibe':494 'vision':123 'visual':154 'voic':298 'voraci':606 'vs':326,921 'want':139,169,177,196,208 'way':93,1165 'week':289,413,416,858,864,1215,1221 'within':831 'without':891,1083,1102,1112 'workflow':430,958 'write':737,1104 'wrong':777,1035 'x':769 'y':774","prices":[{"id":"6b47924e-898b-43f4-8d1a-062cdb9a316f","listingId":"871808bf-8817-4345-9e54-d4c827d85d6f","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"liqiongyu","category":"lenny_skills_plus","install_from":"skills.sh"},"createdAt":"2026-04-18T22:16:58.010Z"}],"sources":[{"listingId":"871808bf-8817-4345-9e54-d4c827d85d6f","source":"github","sourceId":"liqiongyu/lenny_skills_plus/product-taste-intuition","sourceUrl":"https://github.com/liqiongyu/lenny_skills_plus/tree/main/skills/product-taste-intuition","isPrimary":false,"firstSeenAt":"2026-04-18T22:16:58.010Z","lastSeenAt":"2026-04-22T00:56:24.655Z"}],"details":{"listingId":"871808bf-8817-4345-9e54-d4c827d85d6f","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"liqiongyu","slug":"product-taste-intuition","github":{"repo":"liqiongyu/lenny_skills_plus","stars":49,"topics":["agent-skills","ai-agents","automation","claude","codex","prompt-engineering","refoundai","skillpack"],"license":"apache-2.0","html_url":"https://github.com/liqiongyu/lenny_skills_plus","pushed_at":"2026-04-04T06:30:11Z","description":"86 agent-executable skill packs converted from RefoundAI’s Lenny skills (unofficial). Works with Codex + Claude Code.","skill_md_sha":"2a62bf1f0f3691b74f6225ad327719746ebe633c","skill_md_path":"skills/product-taste-intuition/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/liqiongyu/lenny_skills_plus/tree/main/skills/product-taste-intuition"},"layout":"multi","source":"github","category":"lenny_skills_plus","frontmatter":{"name":"product-taste-intuition","description":"Build product taste via a Taste Calibration Sprint (benchmarks, critique notes, hypothesis log)."},"skills_sh_url":"https://skills.sh/liqiongyu/lenny_skills_plus/product-taste-intuition"},"updatedAt":"2026-04-22T00:56:24.655Z"}}