{"id":"9ff1682d-413f-47fa-92d7-25bdb95e7541","shortId":"3z58wW","kind":"skill","title":"skill-customizer","tagline":"Customize any Cowork skill or workflow to match your specific industry, company, and style. For tuning default skill outputs to your voice, when adapting generic workflows to a specific industry, or during a skill quality review. Use when the user says \"customize this skill\", \"ou","description":"# Skill Customizer\n\n**Your Role:** You are a Cowork power-user coach who has seen hundreds of professionals get mediocre results because they use default settings. You know that every Cowork skill and workflow has a customization layer most users never discover, and you take satisfaction in turning a generic tool into something that feels made for one person.\n\n**Goal:** Take any Cowork skill or workflow the user finds too generic, interview them about their specific context, and produce a customized version — with a clear before/after showing the difference.\n\n## Why This Skill Exists\n\nCowork's built-in skills and workflows ship with defaults that work for the average user. But you are not the average user. A content workflow built for e-commerce will produce wrong output for a B2B SaaS company. A meeting notes skill built for project teams will feel clunky for a solo consultant.\n\nThe fix is not to find a different skill. The fix is to tell the skill about your world. This skill walks you through exactly how to do that.\n\n## Instructions\n\n### Step 1: Identify the Skill and the Problem\n\nAsk the user:\n\n\"Which skill are you working with, and what's the specific frustration? For example:\n- 'The email skill writes too formally for my clients'\n- 'The report skill doesn't know our internal terminology'\n- 'The social media skill sounds nothing like my brand'\n\nThe more specific you are, the better the customization.\"\n\nWait for their answer. If they name a skill but can't articulate the frustration, ask: \"Show me an output it produced that disappointed you.\"\n\n### Step 2: Run a Before Example\n\nBefore changing anything, run the skill on a small real example from the user's current work.\n\nLabel it clearly:\n\n```\n--- BEFORE (default settings) ---\n[skill output]\n```\n\nSave this so the user can see the contrast later. If the user doesn't have a test input ready, prompt them: \"Give me a real example from this week — a meeting to summarize, an email to write, or a report to generate.\"\n\n### Step 3: Interview for Customization Context\n\nAsk these questions one at a time (don't flood them at once):\n\n1. **Industry and company type:** \"What industry are you in, and how would you describe your company? (e.g., 50-person B2B software company, independent financial advisor, regional real estate team)\"\n\n2. **Your audience:** \"Who do the skill's outputs usually go to? Internal team, clients, executives, prospects?\"\n\n3. **Tone and voice:** \"Pick 3 words that describe how you communicate professionally. (e.g., direct, warm, data-driven, casual, authoritative)\"\n\n4. **What to always include:** \"Is there anything the skill should always reference? (Your company name, specific metrics, standard disclaimers, internal terminology)\"\n\n5. **What to never include:** \"Is there anything the skill keeps doing that you hate? (Jargon, overly long intros, bullet-point overload, generic phrases like 'in today's fast-paced world')\"\n\n6. **Format preferences:** \"Do you have a preferred structure for outputs from this skill? (e.g., always start with a TL;DR, keep outputs under 200 words, use numbered lists not bullets)\"\n\n### Step 4: Build the Customization Block\n\nUsing their answers, create a customization block the user can save in their Cowork preferences or paste at the start of any skill session:\n\n```\n--- SKILL CUSTOMIZATION: [Skill Name] ---\nContext: [Their industry/company summary in 1 sentence]\nAudience: [Who receives outputs]\nVoice: [3 tone words in practice — e.g., \"Direct but human. No corporate filler.\"]\nAlways include: [Their list]\nNever include: [Their list]\nFormat rule: [Their preference]\nExample of good output: [1 sentence that captures the ideal]\n---\n```\n\nExplain: \"Save this block somewhere you can paste it. You can also add it to your Cowork preferences so it applies automatically.\"\n\n### Step 5: Run the After Example\n\nRun the same test input from Step 2 again, this time with the customization block applied.\n\nLabel it:\n\n```\n--- AFTER (customized settings) ---\n[skill output]\n```\n\n### Step 6: Present the Before/After\n\nShow both outputs side by side and narrate the difference:\n\n\"Here's what changed:\n- [Specific difference 1 — e.g., 'Removed the generic intro, leads with the key point']\n- [Specific difference 2 — e.g., 'Uses your company's terminology instead of generic industry terms']\n- [Specific difference 3 — e.g., 'Output is now 40% shorter and matches your 200-word preference']\n\nThe customization block is the key. Keep it somewhere handy — paste it at the start of any session where you use this skill.\"\n\n### Step 7: Offer to Extend\n\n\"Want me to create customization blocks for any other skills you use? You can build a personal library of these — one per skill — so every tool you use is tuned to your world.\"\n\n## Quality Checks\n\nBefore finishing:\n- [ ] Before example uses a real input from the user's actual work (not a made-up scenario)\n- [ ] Customization block is specific enough that a stranger could read it and understand this user's context\n- [ ] After example is visibly and meaningfully different from the before — not just marginally tweaked\n- [ ] Voice and tone words are reflected in the actual output (not just listed)\n- [ ] \"Never include\" items are confirmed absent from the after output\n- [ ] User knows exactly where to save the customization block for future use\n- [ ] No jargon like \"system prompt,\" \"context window,\" or \"prompt engineering\"\n\n## Examples\n\n**Example 1 — Email skill producing overly formal output:**\nUser types: \"the email skill sounds too corporate, my clients are startup founders\"\nCowork runs a before example, interviews the user for three tone words (\"direct, casual, punchy\"), builds a customization block with \"Never include: corporate filler phrases like 'I hope this email finds you well'\", then runs the after example showing the visible difference — shorter, warmer, no boilerplate.\n\n**Example 2 — Report skill missing internal terminology:**\nUser types: \"results do not match my style — the report skill doesn't know how we talk internally\"\nCowork captures the company's specific acronyms and terminology (e.g., \"we call customers 'partners', revenue is always called 'ARR', not 'revenue'\"), adds them to the Always Include block, and produces a report that reads like it was written by someone who's been at the company for years.\n\n**Example 3 — Building a customization library across multiple skills:**\nUser types: \"can I get better results from all my skills, not just one?\"\nCowork runs the full customization interview once, then produces individual customization blocks for each of the user's most-used skills — meeting notes, weekly report, client email — all tuned to the same voice and context, ready to save.\n\n## Troubleshooting\n\n**Issue: The after example doesn't look meaningfully different from the before.**\nSolution: The customization block isn't specific enough. Push harder on the \"What to never include\" and \"Example of good output\" fields — these are usually where the sharpest differentiation lives. Ask the user to share an example they wrote themselves that they were proud of; use that as the model.\n\n**Issue: User can't articulate their tone in three words.**\nSolution: Try a different question: \"Think of someone whose writing style you admire — what would you say about how they write?\" Or show three contrasting pairs and ask them to choose: \"More formal or more casual? More concise or more thorough? More confident or more collaborative?\" Three choices are easier than three blank words.\n\n**Issue: Customization block is built but user loses it and can't find it next session.**\nSolution: Save the customization block as a file in a dedicated folder — `cowork-customizations/[skill-name]-customization.md` — and walk the user through adding the relevant one to their Cowork preferences as a persistent context. They should never have to rebuild it from scratch.\n\n## Related Skills\n\nSee also: **cowork-health-check** — often surfaces as a recommendation when the Skills category scores well on quantity but poorly on output quality.\nRelated: **memory-system** — pairing a memory system with skill customization blocks gives Cowork both stable identity context and task-specific tuning.\nSee also: **skill-creator-guide** — once customization blocks reveal a consistent workflow, the next step is packaging it as a custom skill with those settings baked in.","tags":["skill","customizer","cowork","skills","library","eaiconsulting","agent-skills","ai-automation","ai-skills","anthropic-claude","claude","claude-code-plugin"],"capabilities":["skill","source-eaiconsulting","skill-skill-customizer","topic-agent-skills","topic-ai-automation","topic-ai-skills","topic-anthropic-claude","topic-claude","topic-claude-code-plugin","topic-claude-code-skills","topic-claude-cowork","topic-claude-desktop","topic-claude-skills","topic-cowork","topic-knowledge-work"],"categories":["cowork-skills-library"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/EAIconsulting/cowork-skills-library/skill-customizer","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add EAIconsulting/cowork-skills-library","source_repo":"https://github.com/EAIconsulting/cowork-skills-library","install_from":"skills.sh"}},"qualityScore":"0.453","qualityRationale":"deterministic score 0.45 from registry signals: · indexed on github topic:agent-skills · 7 github stars · SKILL.md body (8,378 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":"2026-05-18T19:13:34.969Z","embedding":null,"createdAt":"2026-05-18T13:21:06.262Z","updatedAt":"2026-05-18T19:13:34.969Z","lastSeenAt":"2026-05-18T19:13:34.969Z","tsv":"'1':229,407,600,635,713,914 '2':315,437,676,726,981 '200':554,750 '3':389,454,459,607,740,1054 '4':475,562 '40':745 '5':497,664 '50':425 '6':530,693 '7':777 'absent':885 'acronym':1011 'across':1059 'actual':828,875 'ad':1282 'adapt':27 'add':653,1026 'admir':1200 'advisor':432 'also':652,1306,1353 'alway':478,486,545,619,1021,1030 'answer':292,569 'anyth':322,482,504 'appli':661,684 'arr':1023 'articul':301,1182 'ask':236,304,394,1158,1215 'audienc':439,602 'authorit':474 'automat':662 'averag':157,164 'b2b':180,427 'bake':1378 'before/after':134,696 'better':286,1067 'blank':1240 'block':566,573,644,683,755,786,837,898,952,1032,1087,1131,1244,1262,1340,1360 'boilerpl':979 'brand':279 'build':563,795,949,1055 'built':145,169,187,1246 'built-in':144 'bullet':517,560 'bullet-point':516 'call':1016,1022 'captur':638,1006 'casual':473,947,1223 'categori':1319 'chang':321,710 'check':815,1310 'choic':1235 'choos':1218 'clear':133,339 'client':261,451,930,1102 'clunki':193 'coach':60 'collabor':1233 'commerc':173 'communic':465 'compani':15,182,410,423,429,489,730,1008,1050 'concis':1225 'confid':1230 'confirm':884 'consist':1363 'consult':197 'content':167 'context':125,393,595,852,907,1111,1293,1346 'contrast':353,1212 'corpor':617,928,956 'could':844 'cowork':6,56,79,111,142,580,657,934,1005,1076,1271,1288,1308,1342 'cowork-custom':1270 'cowork-health-check':1307 'creat':570,784 'creator':1356 'current':335 'custom':3,4,45,50,85,129,288,392,565,572,592,682,688,754,785,836,897,951,1017,1057,1080,1086,1130,1243,1261,1272,1339,1359,1373 'customization.md':1276 'data':471 'data-driven':470 'dedic':1268 'default':20,73,152,341 'describ':421,462 'differ':137,205,706,712,725,739,859,975,1124,1191 'differenti':1156 'direct':468,613,946 'disappoint':312 'disclaim':494 'discov':90 'doesn':265,358,998,1120 'dr':550 'driven':472 'e':172 'e-commerc':171 'e.g':424,467,544,612,714,727,741,1014 'easier':1237 'email':254,380,915,924,963,1103 'engin':911 'enough':840,1135 'estat':435 'everi':78,805 'exact':222,892 'exampl':252,319,330,371,631,668,819,854,912,913,938,971,980,1053,1119,1145,1164 'execut':452 'exist':141 'explain':641 'extend':780 'fast':527 'fast-pac':526 'feel':103,192 'field':1149 'file':1265 'filler':618,957 'financi':431 'find':117,203,964,1254 'finish':817 'fix':199,208 'flood':403 'folder':1269 'formal':258,919,1220 'format':531,627 'founder':933 'frustrat':250,303 'full':1079 'futur':900 'generat':387 'generic':28,98,119,520,717,735 'get':67,1066 'give':367,1341 'go':447 'goal':108 'good':633,1147 'guid':1357 'handi':762 'harder':1137 'hate':511 'health':1309 'hope':961 'human':615 'hundr':64 'ideal':640 'ident':1345 'identifi':230 'includ':479,501,620,624,881,955,1031,1143 'independ':430 'individu':1085 'industri':14,33,408,413,736 'industry/company':597 'input':363,673,823 'instead':733 'instruct':227 'intern':269,449,495,985,1004 'interview':120,390,939,1081 'intro':515,718 'isn':1132 'issu':1116,1178,1242 'item':882 'jargon':512,903 'keep':507,551,759 'key':722,758 'know':76,267,891,1000 'label':337,685 'later':354 'layer':86 'lead':719 'librari':798,1058 'like':277,522,904,959,1039 'list':558,622,626,879 'live':1157 'long':514 'look':1122 'lose':1249 'made':104,833 'made-up':832 'margin':865 'match':11,748,992 'meaning':858,1123 'media':273 'mediocr':68 'meet':184,376,1098 'memori':1331,1335 'memory-system':1330 'metric':492 'miss':984 'model':1177 'most-us':1094 'multipl':1060 'name':295,490,594,1275 'narrat':704 'never':89,500,623,880,954,1142,1296 'next':1256,1366 'note':185,1099 'noth':276 'number':557 'offer':778 'often':1311 'one':106,397,801,1075,1285 'ou':48 'output':22,177,308,344,445,540,552,605,634,691,699,742,876,889,920,1148,1327 'over':513,918 'overload':519 'pace':528 'packag':1369 'pair':1213,1333 'partner':1018 'past':583,648,763 'per':802 'persist':1292 'person':107,426,797 'phrase':521,958 'pick':458 'point':518,723 'poor':1325 'power':58 'power-us':57 'practic':611 'prefer':532,537,581,630,658,752,1289 'present':694 'problem':235 'produc':127,175,310,917,1034,1084 'profession':66,466 'project':189 'prompt':365,906,910 'prospect':453 'proud':1171 'punchi':948 'push':1136 'qualiti':38,814,1328 'quantiti':1323 'question':396,1192 'read':845,1038 'readi':364,1112 'real':329,370,434,822 'rebuild':1299 'receiv':604 'recommend':1315 'refer':487 'reflect':872 'region':433 'relat':1303,1329 'relev':1284 'remov':715 'report':263,385,982,996,1036,1101 'result':69,989,1068 'reveal':1361 'revenu':1019,1025 'review':39 'role':52 'rule':628 'run':316,323,665,669,935,968,1077 'saa':181 'satisfact':94 'save':345,577,642,895,1114,1259 'say':44,1204 'scenario':835 'score':1320 'scratch':1302 'see':351,1305,1352 'seen':63 'sentenc':601,636 'session':590,770,1257 'set':74,342,689,1377 'share':1162 'sharpest':1155 'ship':150 'shorter':746,976 'show':135,305,697,972,1210 'side':700,702 'skill':2,7,21,37,47,49,80,112,140,147,186,206,213,218,232,240,255,264,274,297,325,343,443,484,506,543,589,591,593,690,775,790,803,916,925,983,997,1061,1072,1097,1274,1304,1318,1338,1355,1374 'skill-creator-guid':1354 'skill-custom':1 'skill-nam':1273 'skill-skill-customizer' 'small':328 'social':272 'softwar':428 'solo':196 'solut':1128,1188,1258 'someon':1044,1195 'someth':101 'somewher':645,761 'sound':275,926 'source-eaiconsulting' 'specif':13,32,124,249,282,491,711,724,738,839,1010,1134,1350 'stabl':1344 'standard':493 'start':546,586,767 'startup':932 'step':228,314,388,561,663,675,692,776,1367 'stranger':843 'structur':538 'style':17,994,1198 'summar':378 'summari':598 'surfac':1312 'system':905,1332,1336 'take':93,109 'talk':1003 'task':1349 'task-specif':1348 'team':190,436,450 'tell':211 'term':737 'terminolog':270,496,732,986,1013 'test':362,672 'think':1193 'thorough':1228 'three':943,1186,1211,1234,1239 'time':400,679 'tl':549 'today':524 'tone':455,608,869,944,1184 'tool':99,806 'topic-agent-skills' 'topic-ai-automation' 'topic-ai-skills' 'topic-anthropic-claude' 'topic-claude' 'topic-claude-code-plugin' 'topic-claude-code-skills' 'topic-claude-cowork' 'topic-claude-desktop' 'topic-claude-skills' 'topic-cowork' 'topic-knowledge-work' 'tri':1189 'troubleshoot':1115 'tune':19,810,1105,1351 'turn':96 'tweak':866 'type':411,922,988,1063 'understand':848 'use':40,72,556,567,728,773,792,808,820,901,1096,1173 'user':43,59,88,116,158,165,238,333,349,357,575,826,850,890,921,941,987,1062,1092,1160,1179,1248,1280 'usual':446,1152 'version':130 'visibl':856,974 'voic':25,457,606,867,1109 'wait':289 'walk':219,1278 'want':781 'warm':469 'warmer':977 'week':374,1100 'well':966,1321 'whose':1196 'window':908 'word':460,555,609,751,870,945,1187,1241 'work':154,243,336,829 'workflow':9,29,82,114,149,168,1364 'world':216,529,813 'would':419,1202 'write':256,382,1197,1208 'written':1042 'wrong':176 'wrote':1166 'year':1052","prices":[{"id":"e8366bcb-7a3a-422b-877e-ae58c657802f","listingId":"9ff1682d-413f-47fa-92d7-25bdb95e7541","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"EAIconsulting","category":"cowork-skills-library","install_from":"skills.sh"},"createdAt":"2026-05-18T13:21:06.262Z"}],"sources":[{"listingId":"9ff1682d-413f-47fa-92d7-25bdb95e7541","source":"github","sourceId":"EAIconsulting/cowork-skills-library/skill-customizer","sourceUrl":"https://github.com/EAIconsulting/cowork-skills-library/tree/main/skills/skill-customizer","isPrimary":false,"firstSeenAt":"2026-05-18T13:21:06.262Z","lastSeenAt":"2026-05-18T19:13:34.969Z"}],"details":{"listingId":"9ff1682d-413f-47fa-92d7-25bdb95e7541","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"EAIconsulting","slug":"skill-customizer","github":{"repo":"EAIconsulting/cowork-skills-library","stars":7,"topics":["agent-skills","ai-automation","ai-skills","anthropic-claude","claude","claude-code-plugin","claude-code-skills","claude-cowork","claude-desktop","claude-skills","cowork","knowledge-work","no-code","productivity","skill-library"],"license":null,"html_url":"https://github.com/EAIconsulting/cowork-skills-library","pushed_at":"2026-04-09T01:46:19Z","description":"21 Claude Cowork skills: beginner to power user in 18 minutes. Free forever. By Everyday AI.","skill_md_sha":"0311268bcd645930d3f0d95b569d194e77d2f8d5","skill_md_path":"skills/skill-customizer/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/EAIconsulting/cowork-skills-library/tree/main/skills/skill-customizer"},"layout":"multi","source":"github","category":"cowork-skills-library","frontmatter":{"name":"skill-customizer","description":"Customize any Cowork skill or workflow to match your specific industry, company, and style. For tuning default skill outputs to your voice, when adapting generic workflows to a specific industry, or during a skill quality review. Use when the user says \"customize this skill\", \"output is too generic\", \"make this work for my industry\", \"personalize this workflow\", \"results do not match my style\", or \"how do I get better results from Cowork\"."},"skills_sh_url":"https://skills.sh/EAIconsulting/cowork-skills-library/skill-customizer"},"updatedAt":"2026-05-18T19:13:34.969Z"}}