{"id":"b142a573-c899-44c4-bb65-3e6898075b7e","shortId":"PdGNSu","kind":"skill","title":"author-strategy","tagline":"PubMed author profile analysis. Author name → PubMed fetch → study type classification → visualization → strategy report.","description":"# /author-strategy — PubMed Author Strategy Analysis\n\n## Purpose\n\nAnalyze a researcher's PubMed publication portfolio to reverse-engineer their research strategy. Produces a CSV dataset, 7 visualizations, and a strategy report.\n\n## Prerequisites\n\n- Python 3.10+ with `biopython`, `pandas`, `matplotlib`, `seaborn`\n- Scripts: `${CLAUDE_SKILL_DIR}/fetch_pubmed.py`, `${CLAUDE_SKILL_DIR}/analyze_patterns.py`\n\n## Workflow\n\n### Step 1: Gather Input\n\nAsk the user for:\n1. **Author name** (PubMed format, e.g., \"Kim DK\" or \"Lee KS\")\n2. **Last name** for position classification (auto-detected if ambiguous)\n3. **Output directory** (default: `~/.local/cache/author-strategy/{AuthorName}/`)\n\n### Step 2: Fetch PubMed Data\n\n```bash\npython \"${CLAUDE_SKILL_DIR}/fetch_pubmed.py\" \"{Author Name}\" \\\n  --last-name \"{LastName}\" \\\n  --output \"{output_dir}/data/{name}_publications.csv\" \\\n  --email \"{user_email}\"\n```\n\nReview the console summary (total count, study type distribution, author position).\nIf count is 0, suggest alternative name formats (e.g., \"Yon DK\" vs \"Yon D\" vs \"Yon Dong Keon\").\n\n### Step 3: Generate Visualizations and Report\n\n```bash\npython \"${CLAUDE_SKILL_DIR}/analyze_patterns.py\" \"{output_dir}/data/{name}_publications.csv\" \\\n  --output-dir \"{output_dir}/report/\" \\\n  --author-name \"{Author Name}\"\n```\n\nThis produces:\n- 7 PNG charts (01-07)\n- `analysis_report.md` with strategy breakdown\n\n### Step 4: Interpret and Present\n\nRead `analysis_report.md` and present to the user:\n\n1. **Executive summary**: total publications, growth trajectory, high-tier rate\n2. **Primary strategy**: what study type dominates and why\n3. **Author position analysis**: leadership rate (1st + last) vs middle\n4. **Topic clusters**: research focus areas\n5. **ROI quadrant**: which strategies yield high-tier + leadership vs. volume only\n6. **Replication opportunities**: which patterns are replicable with Claude Code + public databases\n\n### Step 5: Optional — MA Gap Identification\n\nIf the user asks \"이 교수님과 MA 가능한 주제?\":\n- Cross-reference topic clusters with existing MA plans in memory\n- Identify gaps where the professor has domain expertise but no MA published\n- Output a prioritized list of MA proposals\n\n## Study Type Classifier\n\nThe classifier is tuned for Korean epidemiology and public health researchers. Categories:\n\n| Type | Detection Pattern |\n|------|------------------|\n| GBD | \"global burden\" or \"gbd\" in title/abstract |\n| SR/MA | \"systematic review\" or \"meta-analysis\" |\n| NHIS/Claims | \"national health insurance\", \"nhis\", \"claims database\", \"nationwide cohort\" |\n| Cross-national | Country pairs or \"cross-national\"/\"binational\" |\n| National survey | \"knhanes\", \"nhanes\", \"kchs\", \"national survey\" |\n| Biobank | \"biobank\" |\n| AI/ML | \"machine learning\", \"deep learning\", \"artificial intelligence\" |\n| Clinical trial | \"randomized\" or publication type |\n| Case report | \"case report\" |\n| Letter/Commentary | Publication type = letter/comment/editorial |\n\n**Known limitation**: The classifier may undercount NHIS studies when they appear in Cross-national or Other categories. The report notes this.\n\n## Known Limitations\n\n- The study type classifier is tuned for epidemiology and public health researchers. May undercount specialized study types for other fields.\n- NHIS studies may be undercounted when they appear in cross-national or \"other\" categories.\n- PubMed search requires an email for NCBI E-utilities (set via `--email` flag).\n\n## Anti-Hallucination\n\n- **Never fabricate publication counts, h-index, or journal metrics.** All numbers must come from PubMed API output.\n- **Never invent study classifications.** If a paper cannot be classified, label it as \"Other\" rather than guessing.\n- If PubMed returns 0 results, suggest alternative name formats rather than generating fake data.\n\n## Output Structure\n\n```\n{output_dir}/\n  data/\n    {name}_publications.csv\n  report/\n    analysis_report.md\n    01_yearly_stacked.png\n    02_study_type_pie.png\n    03_author_position.png\n    04_journal_tier_heatmap.png\n    05_topic_distribution.png\n    06_growth_curve.png\n    07_strategy_roi.png\n```","tags":["author","strategy","medsci","skills","aperivue","agent-skills","biostatistics","claude-code","claude-skills","clinical-research","diagnostic-accuracy","irb-protocol"],"capabilities":["skill","source-aperivue","skill-author-strategy","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/author-strategy","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 (4,099 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":"2026-05-18T18:56:28.635Z","embedding":null,"createdAt":"2026-05-13T12:57:43.783Z","updatedAt":"2026-05-18T18:56:28.635Z","lastSeenAt":"2026-05-18T18:56:28.635Z","tsv":"'-07':191 '/.local/cache/author-strategy':100 '/analyze_patterns.py':64,168 '/author-strategy':18 '/data':122,171 '/fetch_pubmed.py':60,112 '/report':179 '0':142,509 '01':190 '01_yearly_stacked.png':529 '02_study_type_pie.png':530 '03_author_position.png':531 '04_journal_tier_heatmap.png':532 '05_topic_distribution.png':533 '06_growth_curve.png':534 '07_strategy_roi.png':535 '1':67,74,208 '1st':234 '2':85,103,219 '3':96,158,228 '3.10':50 '4':197,238 '5':244,270 '6':257 '7':42,187 'ai/ml':374 'altern':144,512 'ambigu':95 'analysi':7,22,231,345 'analysis_report.md':192,202,528 'analyz':24 'anti':469 'anti-hallucin':468 'api':487 'appear':405,446 'area':243 'artifici':379 'ask':70,278 'author':2,5,8,20,75,113,137,181,183,229 'author-nam':180 'author-strategi':1 'authornam':101 'auto':92 'auto-detect':91 'bash':107,163 'binat':364 'biobank':372,373 'biopython':52 'breakdown':195 'burden':334 'cannot':496 'case':387,389 'categori':328,412,453 'chart':189 'claim':351 'classif':14,90,492 'classifi':316,318,398,422,498 'claud':57,61,109,165,265 'clinic':381 'cluster':240,288 'code':266 'cohort':354 'come':484 'consol':130 'count':133,140,474 'countri':358 'cross':285,356,362,408,449 'cross-nat':355,361,407,448 'cross-refer':284 'csv':40 'd':152 'data':106,519,524 'databas':268,352 'dataset':41 'deep':377 'default':99 'detect':93,330 'dir':59,63,111,121,167,170,176,178,523 'directori':98 'distribut':136 'dk':81,149 'domain':301 'domin':225 'dong':155 'e':462 'e-util':461 'e.g':79,147 'email':125,127,458,466 'engin':34 'epidemiolog':323,426 'execut':209 'exist':290 'expertis':302 'fabric':472 'fake':518 'fetch':11,104 'field':438 'flag':467 'focus':242 'format':78,146,514 'gap':273,296 'gather':68 'gbd':332,336 'generat':159,517 'global':333 'growth':213 'guess':505 'h':476 'h-index':475 'hallucin':470 'health':326,348,429 'high':216,251 'high-tier':215,250 'identif':274 'identifi':295 'index':477 'input':69 'insur':349 'intellig':380 'interpret':198 'invent':490 'journal':479 'kchs':369 'keon':156 'kim':80 'knhane':367 'known':395,417 'korean':322 'ks':84 'label':499 'last':86,116,235 'last-nam':115 'lastnam':118 'leadership':232,253 'learn':376,378 'lee':83 'letter/comment/editorial':394 'letter/commentary':391 'limit':396,418 'list':310 'ma':272,281,291,305,312 'machin':375 'matplotlib':54 'may':399,431,441 'memori':294 'meta':344 'meta-analysi':343 'metric':480 'middl':237 'must':483 'name':9,76,87,114,117,123,145,172,182,184,513,525 'nation':347,357,363,365,370,409,450 'nationwid':353 'ncbi':460 'never':471,489 'nhane':368 'nhis':350,401,439 'nhis/claims':346 'note':415 'number':482 'opportun':259 'option':271 'output':97,119,120,169,175,177,307,488,520,522 'output-dir':174 'pair':359 'panda':53 'paper':495 'pattern':261,331 'plan':292 'png':188 'portfolio':30 'posit':89,138,230 'prerequisit':48 'present':200,204 'primari':220 'priorit':309 'produc':38,186 'professor':299 'profil':6 'propos':313 'public':29,212,267,325,385,392,428,473 'publications.csv':124,173,526 'publish':306 'pubm':4,10,19,28,77,105,454,486,507 'purpos':23 'python':49,108,164 'quadrant':246 'random':383 'rate':218,233 'rather':503,515 'read':201 'refer':286 'replic':258,263 'report':17,47,162,388,390,414,527 'requir':456 'research':26,36,241,327,430 'result':510 'return':508 'revers':33 'reverse-engin':32 'review':128,341 'roi':245 'script':56 'seaborn':55 'search':455 'set':464 'skill':58,62,110,166 'skill-author-strategy' 'source-aperivue' 'special':433 'sr/ma':339 'step':66,102,157,196,269 'strategi':3,16,21,37,46,194,221,248 'structur':521 'studi':12,134,223,314,402,420,434,440,491 'suggest':143,511 'summari':131,210 'survey':366,371 'systemat':340 'tier':217,252 'title/abstract':338 'topic':239,287 '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':132,211 'trajectori':214 'trial':382 'tune':320,424 'type':13,135,224,315,329,386,393,421,435 'undercount':400,432,443 'user':72,126,207,277 'util':463 'via':465 'visual':15,43,160 'volum':255 'vs':150,153,236,254 'workflow':65 'yield':249 'yon':148,151,154 '가능한':282 '교수님과':280 '이':279 '주제':283","prices":[{"id":"db06a575-6f36-4310-b5d5-53126bdf6c25","listingId":"b142a573-c899-44c4-bb65-3e6898075b7e","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":"2026-05-13T12:57:43.783Z"}],"sources":[{"listingId":"b142a573-c899-44c4-bb65-3e6898075b7e","source":"github","sourceId":"Aperivue/medsci-skills/author-strategy","sourceUrl":"https://github.com/Aperivue/medsci-skills/tree/main/skills/author-strategy","isPrimary":false,"firstSeenAt":"2026-05-13T12:57:43.783Z","lastSeenAt":"2026-05-18T18:56:28.635Z"}],"details":{"listingId":"b142a573-c899-44c4-bb65-3e6898075b7e","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"Aperivue","slug":"author-strategy","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":"9b907d03f16dea50428188f5ae55e2e9feb694da","skill_md_path":"skills/author-strategy/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/Aperivue/medsci-skills/tree/main/skills/author-strategy"},"layout":"multi","source":"github","category":"medsci-skills","frontmatter":{"name":"author-strategy","description":"PubMed author profile analysis. Author name → PubMed fetch → study type classification → visualization → strategy report."},"skills_sh_url":"https://skills.sh/Aperivue/medsci-skills/author-strategy"},"updatedAt":"2026-05-18T18:56:28.635Z"}}