{"id":"27469cbc-d168-449c-ab5c-a6e99e95e2ff","shortId":"dQeMYc","kind":"skill","title":"grant-builder","tagline":"Grant and challenge proposal support for radiology and medical AI projects. Structures significance, innovation, approach, milestones, and consortium roles while keeping claims evidence-based and executable.","description":"# Grant-Builder Skill\n\n## Purpose\n\nThis skill supports competitive proposal writing for:\n\n- national R&D grants\n- multi-institution consortia\n- challenge proposals\n- internal pilot funding\n- translational medical AI project plans\n- **Korean government grants** (산학협력 / 연구계획서 — MOHW 복지부, MOTIE 산자부, MSS 중기부, and regional industry-academia programs)\n\nIt is optimized for projects where clinical relevance, multi-site coordination, and executable milestones matter as much as technical novelty.\n\n---\n\n## Korean Government Grant Mode (산학과제 / 연구계획서)\n\nWhen the user requests a Korean industry-academia grant (산학과제) or research plan\n(연구계획서), apply the adaptations below. Korean program terms are preserved in\nparentheses because they are the literal form used on the funding agency's template.\n\n### Document Structure (three-attachment format)\n\nMost Korean grants follow a standardized three-attachment format:\n- **Attachment 1 (첨부1, 기본정보)**: project title, participating institutions,\n  investigator CVs, publication / patent record.\n- **Attachment 2 (첨부2, 매칭확인서)**: per-institution cost-share confirmation,\n  typically finalized after a kickoff meeting between the institutions.\n- **Attachment 3 (첨부3, 연구계획서)**: the 10-page research plan — structure below.\n\n### Attachment 3 Standard Structure\n\n```\n1. Significance & Aims (약 2p)\n   - clinical problem with quantitative framing\n   - domestic + international trends (3–5 year literature / guideline window)\n   - differentiation of the proposed work\n\n2. Research Content & Methods (약 4p)\n   - staged roadmap (Phase 1 – N with time ranges)\n   - pipeline schematic (mandatory when an AI pipeline is in scope)\n   - per-subproject institution and personnel assignment\n\n3. Team Capability (약 1p)\n   - expertise + representative record (SCI papers, patents) per investigator\n   - cross-institution synergy (hospital = data / clinical; university = algorithm)\n\n4. Expected Outcomes & Utilization (약 2p)\n   - quantitative targets: SCI papers, patents\n   - qualitative targets: clinical impact, standardization contribution\n   - linkage to follow-on larger grants (positioning as a seed)\n\n5. Budget Plan (약 1p)\n   - RA salaries, computing equipment, consumables, academic activities, indirect costs\n```\n\n### Writing Tips for Small-Scale Grants (< KRW 30 million)\n\n- Write for a non-specialist reviewer — assume the evaluator is not in your subfield.\n- Emphasize feasibility over technical novelty.\n- Prioritize length / format compliance; exceeding the template incurs scoring penalties.\n- Include preliminary data or pilot results whenever available.\n- Keep quantitative targets conservative — undershooting a committed target is punished\n  more than overdelivering on a modest one.\n\n---\n\n## Communication Rules\n\n- Communicate with the user in their preferred language.\n- Proposal prose should be in the language required by the target call.\n- Avoid hype. Emphasize unmet need, feasibility, differentiation, and deliverables.\n\n---\n\n## Core Outputs\n\nDepending on the request, produce one or more of:\n\n- project concept summary\n- `Significance`\n- `Innovation`\n- `Approach`\n- specific aims\n- work packages\n- milestone table\n- role split by institution\n- evaluation framework\n- reviewer-risk memo\n\n---\n\n## Workflow\n\n### Phase 1: Decode the funding call\n\nExtract:\n- funding body\n- call theme\n- eligibility constraints\n- deliverable expectations\n- timeline\n- evaluation criteria\n\nIf no call text is available, infer a generic academic-medical AI proposal structure and label assumptions.\n\n### Phase 2: Frame the problem\n\nDefine:\n- clinical pain point\n- current workflow limitation\n- why existing AI or standard care is insufficient\n- who benefits if the project succeeds\n\n**Gate:** Present the problem framing (clinical pain point, gap, proposed solution) to the\nuser. Confirm before building proposal sections — a misframed problem produces an\nunfundable proposal.\n\n### Phase 3: Build the proposal spine\n\nAlways articulate:\n- problem\n- gap\n- proposed solution\n- why this team can execute it\n- measurable outputs\n\n### Phase 4: Convert to proposal sections\n\n#### Significance\n\nMust answer:\n- why this matters clinically\n- why this matters now\n- why the proposed solution is worth funding\n\n#### Innovation\n\nShould focus on:\n- what is genuinely different\n- why the integration is new\n- why the novelty is useful, not just technical\n\n#### Approach\n\nShould define:\n- dataset and participating sites\n- model or workflow components\n- validation plan\n- benchmark/comparator\n- failure analysis\n- risk mitigation\n\n### Phase 5: Execution plan\n\nGenerate:\n- milestones by quarter or year\n- institution-level responsibilities\n- dependencies and handoffs\n- required infrastructure\n\n---\n\n## Default Structure\n\n```text\n## Proposal Summary\nTitle: ...\nGoal: ...\nClinical problem: ...\n\n### Significance\n...\n\n### Innovation\n...\n\n### Approach\nAim 1. ...\nAim 2. ...\nAim 3. ...\n\n### Milestones\n- ...\n\n### Consortium roles\n- ...\n\n### Major risks and mitigations\n- ...\n```\n\n---\n\n## Evaluation Heuristics\n\nBefore finalizing, check:\n\n1. Is the clinical need explicit and credible?\n2. Is the novelty more than \"we will use AI\"?\n3. Are the aims linked to measurable outputs?\n4. Is the validation plan convincing?\n5. Is the multi-site structure realistic?\n6. Are compute, annotation, and regulatory needs acknowledged?\n7. Does each institution have a distinct role?\n\n---\n\n## Common Weaknesses To Flag\n\n- novelty described without clinical consequence\n- vague benchmark or success criterion\n- no external validation or deployment path\n- too many aims for the timeline\n- consortium members listed but not functionally integrated\n- proposal sounds like a paper, not a funded program\n\n---\n\n## Handoff Rules\n\n- route to `search-lit` to support significance and prior-art positioning\n- route to `design-study` if the evaluation framework is weak\n- route to `write-paper` only when the proposal requires publication-style narrative sections\n\n---\n\n## What This Skill Does NOT Do\n\n- It does not fabricate budget details\n- It does not promise datasets, partners, or infrastructure not evidenced by the user\n- It does not replace institutional administrative review\n\n## Anti-Hallucination\n\n- **Never fabricate references.** All citations must be verified via `/search-lit` with confirmed DOI or PMID. Mark unverified references as `[UNVERIFIED - NEEDS MANUAL CHECK]`.\n- **Never invent clinical definitions, diagnostic criteria, or guideline recommendations.** If uncertain, flag with `[VERIFY]` and ask the user.\n- **Never fabricate numerical results** — compliance percentages, scores, effect sizes, or sample sizes must come from actual data or analysis output.\n- If a reporting guideline item, journal policy, or clinical standard is uncertain, state the uncertainty rather than guessing.","tags":["grant","builder","medsci","skills","aperivue","agent-skills","biostatistics","claude-code","claude-skills","clinical-research","diagnostic-accuracy","irb-protocol"],"capabilities":["skill","source-aperivue","skill-grant-builder","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/grant-builder","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 (6,777 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":"2026-05-18T18:56:29.976Z","embedding":null,"createdAt":"2026-05-13T12:57:44.992Z","updatedAt":"2026-05-18T18:56:29.976Z","lastSeenAt":"2026-05-18T18:56:29.976Z","tsv":"'/search-lit':860 '1':161,208,241,458,660,677 '10':198 '1p':267,317 '2':174,232,494,662,685 '2p':212,290 '3':194,205,221,263,546,664,695 '30':335 '4':285,566,703 '4p':237 '5':222,313,629,709 '6':717 '7':725 'academ':323,485 'academia':76,113 'academic-med':484 'acknowledg':724 'activ':324 'actual':907 'adapt':122 'administr':846 'agenc':141 'ai':13,58,251,487,507,694 'aim':210,441,659,661,663,698,755 'algorithm':284 'alway':551 'analysi':625,910 'annot':720 'answer':573 'anti':849 'anti-hallucin':848 'appli':120 'approach':18,439,610,658 'art':788 'articul':552 'ask':889 'assign':262 'assum':344 'assumpt':492 'attach':148,158,160,173,193,204 'avail':374,480 'avoid':414 'base':28 'benchmark':743 'benchmark/comparator':623 'benefit':514 'bodi':465 'budget':314,826 'build':535,547 'builder':3,33 'call':413,462,466,477 'capabl':265 'care':510 'challeng':6,51 'check':676,873 'citat':855 'claim':25 'clinic':84,213,282,298,499,524,577,654,680,740,876,920 'come':905 'commit':381 'common':733 'communic':392,394 'competit':39 'complianc':360,896 'compon':620 'comput':320,719 'concept':435 'confirm':183,533,862 'consequ':741 'conserv':378 'consortia':50 'consortium':21,666,759 'constraint':469 'consum':322 'content':234 'contribut':301 'convert':567 'convinc':708 'coordin':89 'core':423 'cost':181,326 'cost-shar':180 'credibl':684 'criteria':474,879 'criterion':746 'cross':277 'cross-institut':276 'current':502 'cvs':169 'd':45 'data':281,369,908 'dataset':613,832 'decod':459 'default':647 'defin':498,612 'definit':877 'deliver':422,470 'depend':425,642 'deploy':751 'describ':738 'design':793 'design-studi':792 'detail':827 'diagnost':878 'differ':596 'differenti':227,420 'distinct':731 'document':144 'doi':863 'domest':218 'effect':899 'elig':468 'emphas':352,416 'equip':321 'evalu':346,450,473,672,797 'evid':27 'evidenc':837 'evidence-bas':26 'exceed':361 'execut':30,91,561,630 'exist':506 'expect':286,471 'expertis':268 'explicit':682 'extern':748 'extract':463 'fabric':825,852,893 'failur':624 'feasibl':353,419 'final':185,675 'flag':736,885 'focus':591 'follow':153,305 'follow-on':304 'form':136 'format':149,159,359 'frame':217,495,523 'framework':451,798 'function':764 'fund':55,140,461,464,588,773 'gap':527,554 'gate':519 'generat':632 'generic':483 'genuin':595 'goal':653 'govern':62,100 'grant':2,4,32,46,63,101,114,152,308,333 'grant-build':1,31 'guess':929 'guidelin':225,881,915 'hallucin':850 'handoff':644,775 'heurist':673 'hospit':280 'hype':415 'impact':299 'includ':367 'incur':364 'indirect':325 'industri':75,112 'industry-academia':74,111 'infer':481 'infrastructur':646,835 'innov':17,438,589,657 'institut':49,167,179,192,259,278,449,639,728,845 'institution-level':638 'insuffici':512 'integr':599,765 'intern':53,219 'invent':875 'investig':168,275 'item':916 'journal':917 'keep':24,375 'kickoff':188 'korean':61,99,110,124,151 'krw':334 'label':491 'languag':401,408 'larger':307 'length':358 'level':640 'like':768 'limit':504 'link':699 'linkag':302 'list':761 'lit':781 'liter':135 'literatur':224 'major':668 'mandatori':248 'mani':754 'manual':872 'mark':866 'matter':93,576,580 'measur':563,701 'medic':12,57,486 'meet':189 'member':760 'memo':455 'method':235 'mileston':19,92,444,633,665 'million':336 'misfram':539 'mitig':627,671 'mode':102 'model':617 'modest':390 'mohw':66 'moti':68 'mss':70 'much':95 'multi':48,87,713 'multi-institut':47 'multi-sit':86,712 'must':572,856,904 'n':242 'narrat':814 'nation':43 'need':418,681,723,871 'never':851,874,892 'new':601 'non':341 'non-specialist':340 'novelti':98,356,604,688,737 'numer':894 'one':391,430 'optim':80 'outcom':287 'output':424,564,702,911 'overdeliv':387 'packag':443 'page':199 'pain':500,525 'paper':272,294,770,805 'parenthes':130 'particip':166,615 'partner':833 'patent':171,273,295 'path':752 'penalti':366 'per':178,257,274 'per-institut':177 'per-subproject':256 'percentag':897 'personnel':261 'phase':240,457,493,545,565,628 'pilot':54,371 'pipelin':246,252 'plan':60,118,201,315,622,631,707 'pmid':865 'point':501,526 'polici':918 'posit':309,789 'prefer':400 'preliminari':368 'present':520 'preserv':128 'prior':787 'prior-art':786 'priorit':357 'problem':214,497,522,540,553,655 'produc':429,541 'program':77,125,774 'project':14,59,82,164,434,517 'promis':831 'propos':7,40,52,230,402,488,528,536,544,549,555,569,584,650,766,809 'prose':403 'public':170,812 'publication-styl':811 'punish':384 'purpos':35 'qualit':296 'quantit':216,291,376 'quarter':635 'r':44 'ra':318 'radiolog':10 'rang':245 'rather':927 'realist':716 'recommend':882 'record':172,270 'refer':853,868 'region':73 'regulatori':722 'relev':85 'replac':844 'report':914 'repres':269 'request':108,428 'requir':409,645,810 'research':117,200,233 'respons':641 'result':372,895 'review':343,453,847 'reviewer-risk':452 'risk':454,626,669 'roadmap':239 'role':22,446,667,732 'rout':777,790,801 'rule':393,776 'salari':319 'sampl':902 'scale':332 'schemat':247 'sci':271,293 'scope':255 'score':365,898 'search':780 'search-lit':779 'section':537,570,815 'seed':312 'share':182 'signific':16,209,437,571,656,784 'site':88,616,714 'size':900,903 'skill':34,37,818 'skill-grant-builder' 'small':331 'small-scal':330 'solut':529,556,585 'sound':767 'source-aperivue' 'specialist':342 'specif':440 'spine':550 'split':447 'stage':238 'standard':155,206,300,509,921 'state':924 'structur':15,145,202,207,489,648,715 'studi':794 'style':813 'subfield':351 'subproject':258 'succeed':518 'success':745 'summari':436,651 'support':8,38,783 'synergi':279 'tabl':445 'target':292,297,377,382,412 'team':264,559 'technic':97,355,609 'templat':143,363 'term':126 'text':478,649 'theme':467 'three':147,157 'three-attach':146,156 'time':244 'timelin':472,758 'tip':328 'titl':165,652 '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' 'translat':56 'trend':220 'typic':184 'uncertain':884,923 'uncertainti':926 'undershoot':379 'unfund':543 'univers':283 'unmet':417 'unverifi':867,870 'use':137,606,693 'user':107,397,532,840,891 'util':288 'vagu':742 'valid':621,706,749 'verifi':858,887 'via':859 'weak':734,800 'whenev':373 'window':226 'without':739 'work':231,442 'workflow':456,503,619 'worth':587 'write':41,327,337,804 'write-pap':803 'year':223,637 '기본정보':163 '매칭확인서':176 '복지부':67 '산자부':69 '산학과제':103,115 '산학협력':64 '약':211,236,266,289,316 '연구계획서':65,104,119,196 '중기부':71 '첨부1':162 '첨부2':175 '첨부3':195","prices":[{"id":"63eb0bfc-86b4-49a9-ade9-c129bf07739e","listingId":"27469cbc-d168-449c-ab5c-a6e99e95e2ff","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":"2026-05-13T12:57:44.992Z"}],"sources":[{"listingId":"27469cbc-d168-449c-ab5c-a6e99e95e2ff","source":"github","sourceId":"Aperivue/medsci-skills/grant-builder","sourceUrl":"https://github.com/Aperivue/medsci-skills/tree/main/skills/grant-builder","isPrimary":false,"firstSeenAt":"2026-05-13T12:57:44.992Z","lastSeenAt":"2026-05-18T18:56:29.976Z"}],"details":{"listingId":"27469cbc-d168-449c-ab5c-a6e99e95e2ff","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"Aperivue","slug":"grant-builder","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":"5d81b0b975a3a82e5cc22a95c30faedb2f70ac2e","skill_md_path":"skills/grant-builder/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/Aperivue/medsci-skills/tree/main/skills/grant-builder"},"layout":"multi","source":"github","category":"medsci-skills","frontmatter":{"name":"grant-builder","description":"Grant and challenge proposal support for radiology and medical AI projects. Structures significance, innovation, approach, milestones, and consortium roles while keeping claims evidence-based and executable."},"skills_sh_url":"https://skills.sh/Aperivue/medsci-skills/grant-builder"},"updatedAt":"2026-05-18T18:56:29.976Z"}}