{"id":"a459d8f7-1c4d-428a-a98e-3120e46ef22d","shortId":"rbd2q6","kind":"skill","title":"humanize","tagline":"Detect and remove AI writing patterns from academic manuscripts. Scans for 21 common AI-generated text patterns and rewrites flagged passages to sound naturally human-written while preserving technical accuracy.","description":"# Humanize Skill\n\nYou are assisting a medical researcher in detecting and removing AI writing patterns from\nacademic manuscripts. Your goal: make the text read as if an experienced academic physician\nwrote it, while preserving every technical claim, number, and citation.\n\n## Communication Rules\n\n- Communicate with the user in Korean (matching their working language).\n- All manuscript edits are in English.\n- Medical terminology is always in English, even in Korean communication.\n\n## Reference Files\n\n- **Pattern reference**: `${CLAUDE_SKILL_DIR}/references/ai_patterns.md` -- full 21-pattern list with expanded examples for medical/radiology manuscripts (Pattern 19–21 added 2026-05-01 from senior MA reviewer feedback)\n- **Source material**: Based on matsuikentaro1/humanizer_academic and Wikipedia: Signs of AI writing\n\nAlways read the pattern reference file at the start of a humanize session.\n\n---\n\n## Workflow\n\n### Phase 1: Scan\n\nRead the manuscript section(s) provided by the user and scan for all 21 patterns.\n\n**For each pattern found:**\n1. Record the pattern number and name.\n2. Count occurrences.\n3. Extract the exact passage from the text.\n4. Note the location (paragraph number or line range).\n\n**Output: Pattern Frequency Table**\n\n```\n## AI Pattern Scan Report\n\nSection: {section name}\nWord count: {N}\n\n| # | Pattern | Count | Severity | Example from text |\n|---|---------|-------|----------|-------------------|\n| 1 | Significance inflation | 3 | HIGH | \"...pivotal role in diagnostic imaging...\" |\n| 7 | AI vocabulary words | 5 | HIGH | \"Additionally,...\", \"crucial finding...\" |\n| 8 | Copula avoidance | 2 | MEDIUM | \"...serves as the gold standard...\" |\n| ... | ... | ... | ... | ... |\n\nPatterns not detected: 2, 4, 9, 14, 15\n\nTotal AI pattern instances: {N}\nAI pattern density: {N per 1000 words}\n```\n\n### Phase 2: Report\n\nPresent findings to the user with actionable summary.\n\n**Severity levels:**\n- **HIGH** (>3 occurrences): Likely to trigger AI detection tools. Fix immediately.\n- **MEDIUM** (1-3 occurrences): Noticeable to careful readers. Should fix.\n- **LOW** (0 occurrences): Clean for this pattern.\n\n**AI Pattern Score:**\n- Count total pattern instances across all 21 categories.\n- Compute density: instances per 1000 words.\n- Target: < 2.0 instances per 1000 words.\n\n**Gate:** Present the report and ask the user which patterns to fix. Default: fix all HIGH and MEDIUM.\n\n### Phase 3: Fix\n\nRewrite flagged passages following these rules:\n\n1. **Preserve technical accuracy.** Every number, statistic, p-value, confidence interval, and\n   clinical fact must remain identical.\n2. **Preserve citation density.** Do not remove or relocate citations.\n3. **Preserve formal academic register.** Do not make the text casual or conversational.\n4. **Do not force casualness.** The target voice is an experienced radiologist writing for peers\n   in a top-tier journal -- not a blog post.\n5. **Keep domain-specific terminology intact.** \"Convolutional neural network,\" \"apparent diffusion\n   coefficient,\" \"Fleiss' kappa\" stay as-is.\n6. **Never introduce new claims** or remove existing ones.\n7. **Vary sentence structure.** Mix short declarative sentences (8-12 words) with longer ones\n   (25-35 words). Avoid uniform length.\n8. **Use active voice** where natural. \"We analyzed\" rather than \"Analysis was performed.\"\n\n**Fix strategies per pattern category:**\n\n| Category | Strategy |\n|----------|----------|\n| Content patterns (1-6) | Delete vague claims; replace with specific data or citations |\n| Language patterns (7-12) | Substitute with plain academic English; simplify verb constructions |\n| Style patterns (13-15) | Adjust formatting and punctuation |\n| Filler and hedging (16-18) | Delete filler; calibrate hedging to match evidence level |\n\n**Output:** Present the rewritten text with changes highlighted using diff format or tracked changes.\n\n### Phase 4: Verify\n\nRe-scan the rewritten text using the same 21 patterns.\n\n**Output: Verification Report**\n\n```\n## Verification Report\n\n| Metric | Before | After |\n|--------|--------|-------|\n| Total instances | 23 | 4 |\n| Density (per 1000 words) | 8.2 | 1.4 |\n| HIGH severity patterns | 3 | 0 |\n| MEDIUM severity patterns | 5 | 2 |\n\nRemaining issues:\n- Pattern 17 (hedging): 2 instances remain -- appropriate for the evidence level.\n\nVerdict: PASS (density < 2.0)\n```\n\nIf the density remains above 2.0, run another fix-verify cycle (max 3 rounds).\n\n---\n\n## The 21 Detection Patterns\n\n### Content Patterns\n\n| # | Pattern | What to look for | Fix |\n|---|---------|------------------|-----|\n| 1 | Significance inflation | \"pivotal,\" \"evolving landscape,\" \"underscores the critical importance\" | Delete or state the specific importance with data |\n| 2 | Notability claims | \"landmark trial,\" \"renowned investigators,\" \"groundbreaking\" | Remove; let the data speak |\n| 3 | Superficial -ing analyses | \"highlighting the cardioprotective effects,\" \"underscoring the broad applicability\" | End the sentence at the data; start a new sentence for interpretation |\n| 4 | Promotional language | \"remarkable findings,\" \"dramatic reductions,\" \"profound impact\" | State the actual numbers neutrally |\n| 5 | Vague attributions | \"Studies have shown,\" \"Experts argue,\" \"Several publications\" | Cite the specific study |\n| 6 | Formulaic challenges sections | \"Despite challenges... future outlook... continues to provide\" | State specific limitations factually |\n\n### Language Patterns\n\n| # | Pattern | What to look for | Fix |\n|---|---------|------------------|-----|\n| 7 | AI vocabulary words | Additionally, crucial, delve, enhance, fostering, pivotal, showcase, tapestry, underscore, landscape (abstract) | Delete or replace with plain English |\n| 8 | Copula avoidance | \"serves as,\" \"stands as,\" \"represents a\" | Use \"is\" |\n| 9 | Negative parallelisms | \"not only X but also Y\" | \"X and Y\" |\n| 10 | Rule of three overuse | Forcing ideas into groups of three repeatedly | Use natural grouping (2, 4, 5 items) |\n| 11 | Synonym cycling | patients/participants/subjects/individuals | Pick one term, use consistently |\n| 12 | False ranges | \"from improved renal function to enhanced cardiac outcomes\" | List the specific outcomes directly |\n\n### Style Patterns\n\n| # | Pattern | What to look for | Fix |\n|---|---------|------------------|-----|\n| 13 | Em dash overuse | More than 2 em dashes per page | Use parentheses or restructure |\n| 14 | Title case in headings | \"Statistical Analysis And Primary Endpoints\" | Sentence case per journal style |\n| 15 | Curly quotation marks | Curly quotes from ChatGPT | Straight quotes |\n\n### Filler and Hedging\n\n| # | Pattern | What to look for | Fix |\n|---|---------|------------------|-----|\n| 16 | Filler phrases | \"It is important to note that,\" \"In order to,\" \"Due to the fact that\" | Delete the filler; state the content directly |\n| 17 | Excessive hedging | \"may potentially suggest the possibility\" | Choose the appropriate certainty level: \"suggests\" |\n| 18 | Generic positive conclusions | \"The future looks bright,\" \"continues to reshape,\" \"paves the way\" | State the specific next step or implication |\n\n### Senior MA Reviewer Patterns\n\n| # | Pattern | What to look for | Fix |\n|---|---------|------------------|-----|\n| 19 | § (section sign) marker | \"as in §2.3\", \"(see §Discussion)\", \"§Results\" | Delete or replace with section name (\"Methods\", \"Results\") — `grep -c \"§\"` = 0 |\n| 20 | Methods/Results self-reference parenthetical | \"(Methods §X)\", \"(Results §3.1)\", \"(Methods, Section 2.3)\" | Drop the parenthetical or shorten to \"(see Methods)\" |\n| 21 | AI Disclosure boilerplate (body) | \"## Artificial Intelligence Disclosure\", \"Generative AI was not used to create...\" in manuscript body | Remove from body → place in cover letter / submission form only (per `~/.claude/rules/journal-ai-image-policies.md`) |\n\n---\n\n## Section-Specific Focus\n\nWhen scanning a full manuscript, prioritize these patterns per section:\n\n| Section | Priority Patterns | Reason |\n|---------|------------------|--------|\n| Abstract | ALL (1-21) | Most visible section; most scrutinized for AI patterns |\n| Introduction | 1, 2, 5, 7, 12 | AI inflates background importance and uses vague attributions |\n| Methods | 8, 16 | Methods should be straightforward; copula avoidance and filler are common |\n| Results | 3, 4, 6, 10, 11 | AI adds interpretive -ing clauses and promotional language to results |\n| Discussion | 1, 5, 6, 17, 18 | AI produces formulaic discussions with excessive hedging |\n| Conclusion | 1, 18 | AI generates generic positive conclusions |\n| Methods (MA / SR) | 19, 20, 21 | § markers, self-reference parentheticals, AI Disclosure boilerplate are senior-MA-reviewer red flags |\n| Discussion (MA / SR) | 19, 20 | Self-reference parentheticals especially common when discussing methods |\n| Body (any) | 21 | AI Disclosure belongs in cover letter / submission form, not manuscript body |\n\n---\n\n## Interaction with Other Skills\n\n| Calling skill | When this skill is invoked |\n|---------------|---------------------------|\n| `/write-paper` | Phase 7 (Polish) -- automatic scan before submission |\n| `/peer-review` | When reviewing one's own manuscript for AI patterns |\n\nWhen called by another skill, return the verification report so the calling skill can check\nthe pass/fail status.\n\n---\n\n## What This Skill Does NOT Do\n\n- Does not evaluate scientific quality, accuracy, or completeness of the manuscript.\n- Does not add new content or citations.\n- Does not assess journal compliance or formatting.\n- Does not translate between languages.\n- Only removes AI patterns; does not perform general copy-editing.\n\n## Anti-Hallucination\n\n- **Never introduce new claims or citations** during rewriting. Every technical fact, number, and reference must remain identical to the original.\n- **Never remove existing citations** or relocate them during pattern fixes.\n- **Never change the meaning** of a sentence while fixing AI patterns — only rephrase, never reinterpret.\n- If a passage cannot be fixed without changing its meaning, flag it for the user rather than guessing.\n\n---\n\n## Gates\n\n| Gate | Severity | Trigger | Action on fail |\n|---|---|---|---|\n| AI-pattern density target | ADVISORY | density > 2.0 patterns / 1000 words after sweep | warn; surface remaining flagged passages for manual review |\n| Pattern 19 — `§` symbol | ENFORCED (senior MA reviewer prep) | `grep -c \"§\" manuscript.md` > 0 | auto-strip; verify post-rewrite count == 0 |\n| Pattern 20 — `(see Methods §X)` self-reference | ENFORCED | match found | rewrite to direct section name reference |\n| Pattern 21 — AI Disclosure paragraph in body | ENFORCED | \"Generative AI was not used...\" paragraph in manuscript body | move to cover letter or remove |\n| Citation preservation invariant | ENFORCED | any pre-existing `[@bibkey]` removed by rewrite | revert that single rewrite; flag for user |\n| Numerical preservation invariant | ENFORCED | any number changed by rewrite | revert; flag for user |","tags":["humanize","medsci","skills","aperivue","agent-skills","biostatistics","claude-code","claude-skills","clinical-research","diagnostic-accuracy","irb-protocol","literature-review"],"capabilities":["skill","source-aperivue","skill-humanize","topic-agent-skills","topic-biostatistics","topic-claude-code","topic-claude-skills","topic-clinical-research","topic-diagnostic-accuracy","topic-irb-protocol","topic-literature-review","topic-manuscript","topic-medical-ai","topic-medical-research","topic-meta-analysis"],"categories":["medsci-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/Aperivue/medsci-skills/humanize","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add Aperivue/medsci-skills","source_repo":"https://github.com/Aperivue/medsci-skills","install_from":"skills.sh"}},"qualityScore":"0.499","qualityRationale":"deterministic score 0.50 from registry signals: · indexed on github topic:agent-skills · 98 github stars · SKILL.md body (10,968 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":"2026-05-18T18:56:30.068Z","embedding":null,"createdAt":"2026-05-13T12:57:45.080Z","updatedAt":"2026-05-18T18:56:30.068Z","lastSeenAt":"2026-05-18T18:56:30.068Z","tsv":"'-01':126 '-05':125 '-12':469,516 '-15':528 '-18':537 '-21':1059 '-3':301 '-35':475 '-6':503 '/.claude/rules/journal-ai-image-policies.md':1037 '/peer-review':1200 '/references/ai_patterns.md':109 '/write-paper':1192 '0':310,596,986,1380,1389 '1':158,179,226,300,366,502,646,1058,1069,1112,1125 '1.4':591 '10':796,1099 '1000':273,331,337,588,1357 '11':815,1100 '12':824,1073 '13':527,848 '14':261,863 '15':262,878 '16':536,897,1084 '17':605,921,1115 '18':935,1116,1126 '19':121,966,1135,1156,1370 '2':186,248,258,276,384,601,607,664,811,854,1070 '2.0':334,618,624,1355 '2.3':972,999 '20':987,1136,1157,1391 '2026':124 '21':13,111,122,173,325,572,635,1008,1137,1169,1408 '23':584 '25':474 '3':189,229,289,358,394,595,632,677,1096 '3.1':996 '4':197,259,407,561,585,701,812,1097 '5':240,432,600,715,813,1071,1113 '6':451,729,1098,1114 '7':236,460,515,752,1072,1194 '8':245,468,480,773,1083 '8.2':590 '9':260,784 'abstract':766,1056 'academ':9,50,62,397,520 'accuraci':33,369,1239 'across':323 'action':284,1345 'activ':482 'actual':712 'ad':123 'add':1102,1247 'addit':242,756 'adjust':529 'advisori':1353 'ai':5,16,46,141,210,237,264,268,294,316,753,1009,1017,1066,1074,1101,1117,1127,1143,1170,1208,1266,1317,1349,1409,1416 'ai-gener':15 'ai-pattern':1348 'also':791 'alway':95,143 'analys':680 'analysi':490,869 'analyz':487 'anoth':626,1213 'anti':1276 'anti-hallucin':1275 'appar':442 'applic':688 'appropri':610,931 'argu':722 'artifici':1013 'as-i':448 'ask':344 'assess':1254 'assist':38 'attribut':717,1081 'auto':1382 'auto-strip':1381 'automat':1196 'avoid':247,477,775,1090 'background':1076 'base':134 'belong':1172 'bibkey':1438 'blog':430 'bodi':1012,1025,1028,1167,1180,1413,1423 'boilerpl':1011,1145 'bright':942 'broad':687 'c':985,1378 'calibr':540 'call':1185,1211,1221 'cannot':1326 'cardiac':833 'cardioprotect':683 'care':305 'case':865,874 'casual':404,411 'categori':326,497,498 'certainti':932 'challeng':731,734 'chang':552,559,1309,1330,1455 'chatgpt':885 'check':1224 'choos':929 'citat':73,386,393,512,1251,1283,1301,1430 'cite':725 'claim':70,455,506,666,1281 'claud':106 'claus':1105 'clean':312 'clinic':379 'coeffici':444 'common':14,1094,1163 'communic':74,76,101 'complet':1241 'complianc':1256 'comput':327 'conclus':938,1124,1131 'confid':376 'consist':823 'construct':524 'content':500,638,919,1249 'continu':737,943 'convers':406 'convolut':439 'copi':1273 'copula':246,774,1089 'copy-edit':1272 'count':187,218,221,319,1388 'cover':1031,1174,1426 'creat':1022 'critic':654 'crucial':243,757 'cur':879,882 'cycl':630,817 'dash':850,856 'data':510,663,675,694 'declar':466 'default':351 'delet':504,538,656,767,914,976 'delv':758 'densiti':270,328,387,586,617,621,1351,1354 'despit':733 'detect':2,43,257,295,636 'diagnost':234 'diff':555 'diffus':443 'dir':108 'direct':839,920,1403 'disclosur':1010,1015,1144,1171,1410 'discuss':974,1111,1120,1153,1165 'domain':435 'domain-specif':434 'dramat':706 'drop':1000 'due':909 'edit':88,1274 'effect':684 'em':849,855 'end':689 'endpoint':872 'enforc':1372,1398,1414,1433,1452 'english':91,97,521,772 'enhanc':759,832 'especi':1162 'evalu':1236 'even':98 'everi':68,370,1286 'evid':544,613 'evolv':650 'exact':192 'exampl':116,223 'excess':922,1122 'exist':458,1300,1437 'expand':115 'experienc':61,417 'expert':721 'extract':190 'fact':380,912,1288 'factual':743 'fail':1347 'fals':825 'feedback':131 'file':103,148 'filler':533,539,888,898,916,1092 'find':244,279,705 'fix':297,308,350,352,359,493,628,645,751,847,896,965,1307,1316,1328 'fix-verifi':627 'flag':22,361,1152,1333,1364,1446,1459 'fleiss':445 'focus':1041 'follow':363 'forc':410,801 'form':1034,1177 'formal':396 'format':530,556,1258 'formula':730,1119 'foster':760 'found':178,1400 'frequenc':208 'full':110,1045 'function':830 'futur':735,940 'gate':339,1341,1342 'general':1271 'generat':17,1016,1128,1415 'generic':936,1129 'goal':53 'gold':253 'grep':984,1377 'groundbreak':671 'group':804,810 'guess':1340 'hallucin':1277 'head':867 'hedg':535,541,606,890,923,1123 'high':230,241,288,354,592 'highlight':553,681 'human':1,28,34,154 'human-written':27 'idea':802 'ident':383,1294 'imag':235 'immedi':298 'impact':709 'implic':955 'import':655,661,902,1077 'improv':828 'inflat':228,648,1075 'ing':679,1104 'instanc':266,322,329,335,583,608 'intact':438 'intellig':1014 'interact':1181 'interpret':700,1103 'interv':377 'introduc':453,1279 'introduct':1068 'invari':1432,1451 'investig':670 'invok':1191 'issu':603 'item':814 'journal':427,876,1255 'kappa':446 'keep':433 'korean':81,100 'landmark':667 'landscap':651,765 'languag':85,513,703,744,1108,1263 'length':479 'let':673 'letter':1032,1175,1427 'level':287,545,614,933 'like':291 'limit':742 'line':204 'list':113,835 'locat':200 'longer':472 'look':643,749,845,894,941,963 'low':309 'ma':129,957,1133,1149,1154,1374 'make':54,401 'manual':1367 'manuscript':10,51,87,119,162,1024,1046,1179,1206,1244,1422 'manuscript.md':1379 'mark':881 'marker':969,1138 'match':82,543,1399 'materi':133 'matsuikentaro1/humanizer_academic':136 'max':631 'may':924 'mean':1311,1332 'medic':40,92 'medical/radiology':118 'medium':249,299,356,597 'method':982,993,997,1007,1082,1085,1132,1166,1393 'methods/results':988 'metric':579 'mix':464 'move':1424 'must':381,1292 'n':219,267,271 'name':185,216,981,1405 'natur':26,485,809 'negat':785 'network':441 'neural':440 'neutral':714 'never':452,1278,1298,1308,1321 'new':454,697,1248,1280 'next':952 'notabl':665 'note':198,904 'notic':303 'number':71,183,202,371,713,1289,1454 'numer':1449 'occurr':188,290,302,311 'one':459,473,820,1203 'order':907 'origin':1297 'outcom':834,838 'outlook':736 'output':206,546,574 'overus':800,851 'p':374 'p-valu':373 'page':858 'paragraph':201,1411,1420 'parallel':786 'parenthes':860 'parenthet':992,1002,1142,1161 'pass':616 'pass/fail':1226 'passag':23,193,362,1325,1365 'patients/participants/subjects/individuals':818 'pattern':7,19,48,104,112,120,146,174,177,182,207,211,220,255,265,269,315,317,321,348,496,501,514,526,573,594,599,604,637,639,640,745,746,841,842,891,959,960,1049,1054,1067,1209,1267,1306,1318,1350,1356,1369,1390,1407 'pave':946 'peer':421 'per':272,330,336,495,587,857,875,1036,1050 'perform':492,1270 'phase':157,275,357,560,1193 'phrase':899 'physician':63 'pick':819 'pivot':231,649,761 'place':1029 'plain':519,771 'polish':1195 'posit':937,1130 'possibl':928 'post':431,1386 'post-rewrit':1385 'potenti':925 'pre':1436 'pre-exist':1435 'prep':1376 'present':278,340,547 'preserv':31,67,367,385,395,1431,1450 'primari':871 'priorit':1047 'prioriti':1053 'produc':1118 'profound':708 'promot':702,1107 'provid':165,739 'public':724 'punctuat':532 'qualiti':1238 'quot':883,887 'quotat':880 'radiologist':418 'rang':205,826 'rather':488,1338 're':564 're-scan':563 'read':57,144,160 'reader':306 'reason':1055 'record':180 'red':1151 'reduct':707 'refer':102,105,147,991,1141,1160,1291,1397,1406 'regist':398 'reinterpret':1322 'reloc':392,1303 'remain':382,602,609,622,1293,1363 'remark':704 'remov':4,45,390,457,672,1026,1265,1299,1429,1439 'renal':829 'renown':669 'repeat':807 'rephras':1320 'replac':507,769,978 'report':213,277,342,576,578,1218 'repres':780 'research':41 'reshap':945 'restructur':862 'result':975,983,995,1095,1110 'return':1215 'revert':1442,1458 'review':130,958,1150,1202,1368,1375 'rewrit':21,360,1285,1387,1401,1441,1445,1457 'rewritten':549,567 'role':232 'round':633 'rule':75,365,797 'run':625 'scan':11,159,170,212,565,1043,1197 'scientif':1237 'score':318 'scrutin':1064 'section':163,214,215,732,967,980,998,1039,1051,1052,1062,1404 'section-specif':1038 'see':973,1006,1392 'self':990,1140,1159,1396 'self-refer':989,1139,1158,1395 'senior':128,956,1148,1373 'senior-ma-review':1147 'sentenc':462,467,691,698,873,1314 'serv':250,776 'session':155 'sever':222,286,593,598,723,1343 'short':465 'shorten':1004 'showcas':762 'shown':720 'sign':139,968 'signific':227,647 'simplifi':522 'singl':1444 'skill':35,107,1184,1186,1189,1214,1222,1230 'skill-humanize' 'sound':25 'sourc':132 'source-aperivue' 'speak':676 'specif':436,509,660,727,741,837,951,1040 'sr':1134,1155 'stand':778 'standard':254 'start':151,695 'state':658,710,740,917,949 'statist':372,868 'status':1227 'stay':447 'step':953 'straight':886 'straightforward':1088 'strategi':494,499 'strip':1383 'structur':463 'studi':718,728 'style':525,840,877 'submiss':1033,1176,1199 'substitut':517 'suggest':926,934 'summari':285 'superfici':678 'surfac':1362 'sweep':1360 'symbol':1371 'synonym':816 'tabl':209 'tapestri':763 'target':333,413,1352 'technic':32,69,368,1287 'term':821 'terminolog':93,437 'text':18,56,196,225,403,550,568 'three':799,806 'tier':426 'titl':864 'tool':296 'top':425 'top-tier':424 'topic-agent-skills' 'topic-biostatistics' 'topic-claude-code' 'topic-claude-skills' 'topic-clinical-research' 'topic-diagnostic-accuracy' 'topic-irb-protocol' 'topic-literature-review' 'topic-manuscript' 'topic-medical-ai' 'topic-medical-research' 'topic-meta-analysis' 'total':263,320,582 'track':558 'translat':1261 'trial':668 'trigger':293,1344 'underscor':652,685,764 'uniform':478 'use':481,554,569,782,808,822,859,1020,1079,1419 'user':79,168,282,346,1337,1448,1461 'vagu':505,716,1080 'valu':375 'vari':461 'verb':523 'verdict':615 'verif':575,577,1217 'verifi':562,629,1384 'visibl':1061 'vocabulari':238,754 'voic':414,483 'warn':1361 'way':948 'wikipedia':138 'without':1329 'word':217,239,274,332,338,470,476,589,755,1358 'work':84 'workflow':156 'write':6,47,142,419 'written':29 'wrote':64 'x':789,793,994,1394 'y':792,795","prices":[{"id":"602a80ad-ea2b-4277-8ea5-cf83ee6ab233","listingId":"a459d8f7-1c4d-428a-a98e-3120e46ef22d","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"Aperivue","category":"medsci-skills","install_from":"skills.sh"},"createdAt":"2026-05-13T12:57:45.080Z"}],"sources":[{"listingId":"a459d8f7-1c4d-428a-a98e-3120e46ef22d","source":"github","sourceId":"Aperivue/medsci-skills/humanize","sourceUrl":"https://github.com/Aperivue/medsci-skills/tree/main/skills/humanize","isPrimary":false,"firstSeenAt":"2026-05-13T12:57:45.080Z","lastSeenAt":"2026-05-18T18:56:30.068Z"}],"details":{"listingId":"a459d8f7-1c4d-428a-a98e-3120e46ef22d","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"Aperivue","slug":"humanize","github":{"repo":"Aperivue/medsci-skills","stars":98,"topics":["agent-skills","biostatistics","claude-code","claude-skills","clinical-research","diagnostic-accuracy","irb-protocol","literature-review","manuscript","medical-ai","medical-research","meta-analysis","physician-researcher","prisma","pubmed","radiology","reporting-guidelines","strobe","systematic-review","tripod-ai"],"license":"other","html_url":"https://github.com/Aperivue/medsci-skills","pushed_at":"2026-05-17T20:50:52Z","description":"Claude Code skills for medical research — literature search, reporting guidelines, statistical analysis, publication figures. Built by a physician-researcher, tested on real publications. MIT licensed.","skill_md_sha":"aead2e00a2118c0a1b31187f472429afba87ed17","skill_md_path":"skills/humanize/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/Aperivue/medsci-skills/tree/main/skills/humanize"},"layout":"multi","source":"github","category":"medsci-skills","frontmatter":{"name":"humanize","description":"Detect and remove AI writing patterns from academic manuscripts. Scans for 21 common AI-generated text patterns and rewrites flagged passages to sound naturally human-written while preserving technical accuracy."},"skills_sh_url":"https://skills.sh/Aperivue/medsci-skills/humanize"},"updatedAt":"2026-05-18T18:56:30.068Z"}}