[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71QQ82F70RMFMF066T8JM34":3,"skill-same:01M71QQ82F70RMFMF066T8JM34":65},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":37,"linkOnly":38,"markdownContent":39,"markdownBody":40,"frontMatter":41,"refName":42,"commitSha":43,"skillDirectory":44,"forksCount":45,"repositoryHtmlUrl":46,"repositoryHomepageUrl":47,"ownerHtmlUrl":48,"ownerBlogUrl":6,"ownerDescription":49,"updatedAt":50,"install":51,"resources":64},"01M71QQ82F70RMFMF066T8JM34","method-voice",null,"Applies a precise, clinically detached voice to Introductions, Discussions, abstracts, cover letters, and grants. Use when the user wants that voice, tone, or hedging; full manuscript drafting goes to method-write.","sajor2000\u002Fmethodsmith","sajor2000","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F126191042?v=4","01M71QG4M3561SH49E3S7J3GKT","JC Rojas ","01M71QG4N56X2P61YTJSRP95VD",6,"MIT","https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Fplugins\u002Fmethodsmith\u002Fskills\u002Fmethod-voice\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36],"method voice","voice","manuscript","use","when","drafting","tone","grants","write","cover","full","user","goes","2026-11-10 13:02:02",false,"---\nname: method-voice\ndescription: \"Applies a precise, clinically detached voice to Introductions, Discussions, abstracts, cover letters, and grants. Use when the user wants that voice, tone, or hedging; full manuscript drafting goes to method-write.\"\n---\n\n# Clinical Research Voice\n\n## Skill Value\n\n- **Problem it solves:** Clinical AI prose can blur evidence strength, authorial responsibility, and the boundary between narrative interpretation and factual reporting.\n- **Use when:** Drafting or revising an Introduction, Discussion, narrative abstract text, cover letter, grant text, or related scientific prose.\n- **Output:** Evidence-calibrated, clinically detached prose with a focused voice self-check.\n- **Ask only if:** The manuscript section or intended scientific register cannot be inferred from the supplied text.\n- **Do not do:** Do not apply interpretive voice rules to Methods or Results, and do not invent evidence or citations.\n\nHouse style for formal scientific reporting in clinical AI research: precise, methodologically transparent, hedged only where evidence warrants, direct where findings are clear. Every claim traces to data or citation. The quoted sentences and statistics in this skill illustrate sentence form only; they are not sourced findings and must not be reused as evidence or citations.\n\n**Scope.** This voice governs narrative prose — Abstract background\u002Fconclusions, Introduction, Discussion, cover letters, grant narrative. It does not govern Methods or Results, which are audited separately under `method-section-check` and must stay purely factual and procedural. Applying this skill's interpretive latitude (hedging, mechanism, implications) to Methods or Results is itself an error — check which section you're in before applying these rules.\n\n## Sentence Architecture\n\nOpen sentences with the grammatical subject, then the verb, then qualification: \"Physicians demonstrate substantial automation bias when exposed to erroneous LLM recommendations, even with voluntary consultation and prior AI literacy training.\" Avoid front-loaded rhetorical flourishes. State the point, then elaborate — do not build toward a conclusion.\n\nVary sentence length deliberately: short declarative anchors (\"Most diagnostic errors arise from judgment-related pitfalls\") alongside complex subordinated sentences of 40+ words that specify population, condition, and effect size in a single unit.\n\nWithin a paragraph, sequence: (1) the finding or problem, (2) mechanism or context, (3) connection to implications or prior literature. A prior-literature statement carries a citation to a source that was read and resolves to a PMID or DOI; obtain it through `method-write`'s evidence access or `method-citations`, never from memory.\n\n## Pronouns and Point of View\n\nUse \"we\" for authorial actions and interpretation: \"we conducted,\" \"we find,\" \"we evaluated,\" \"we interpret this as.\" Reserve passive constructions (\"participants were randomly assigned,\" \"scoring rubrics were created\") for procedural steps where the actor is the study apparatus, not the authors — this passive register belongs primarily to Methods, not to Discussion prose.\n\nNever use \"I.\" Address readers or future researchers as implied subjects (\"Future research should explore\") rather than as \"you.\"\n\n## Punctuation Rules\n\nUse em-dashes to embed a definition mid-sentence: \"automation bias — the tendency to over-rely on automated output.\" Definitional embedding is the sanctioned use in this house style. Non-definitional em-dashes (for suspense, rhythm, or as a comma substitute) are a tell — rewrite them with commas, semicolons, or parentheticals. High em-dash density is itself a detector signal even when each use is definitional; if a paragraph accumulates several, convert all but the most load-bearing definition to parenthetical or appositive form.\n\nUse commas and semicolons to link closely related independent clauses. Avoid ellipses. Use parentheticals for statistics: \"(95% CI, −18.9 to −9.1; P\u003C.001).\" Format P values to the target journal's convention (leading zero, decimal places, and the lowest reported bound, commonly P\u003C.001 or P\u003C0.001); do not report P\u003C0.0001 unless the journal allows it.\n\nRhetorical questions are permitted only to frame an evidence gap — \"Are AI-trained physicians exercising voluntary consultation vulnerable to automation bias when LLM recommendations contain errors?\" — and must be followed immediately by the methodological response. Never leave one dangling.\n\n## Vocabulary Instructions\n\nPrefer precise technical nouns over informal approximations: \"hallucinate,\" \"automation bias,\" \"intraclass correlation coefficient,\" \"intention-to-treat,\" \"clinical vignettes.\"\n\nCalibrate verbs to strength of evidence: \"demonstrate\" for strong evidence, \"indicate\"\u002F\"reveal\" for direct findings, \"suggest\" for exploratory findings. Reserve \"significant\" for the result of a prespecified hypothesis test, reported with its estimate and confidence interval; never use it as a synonym for \"large\" or \"important.\" Use \"substantial\" only alongside an effect size and interval, and \"robust\" only when named sensitivity analyses agree (ASA statement on P values, Am Stat 2016;70:129-33, principles 3 and 5). Apply hedging phrases to subgroup and secondary analyses specifically: \"should be regarded as hypothesis-generating rather than conclusive.\"\n\nDraw analogies from clinical practice and cognitive psychology where they clarify mechanism: \"illness scripts,\" \"cognitive offloading,\" \"anchoring,\" \"fluency-induced heuristic processing.\" Analogy is for the Discussion's mechanistic explanation, not for Methods or Results.\n\n## Tone and Emotional Range\n\nMaintain clinical detachment throughout. State alarming findings plainly, without amplification: \"raising important patient safety concerns for clinical integration.\" Acknowledge limitations in a dedicated Discussion subsection using neutral concessive constructions: \"Several limitations of this study warrant consideration.\" Do not editorialize.\n\nWhen findings challenge assumptions, say so directly and follow immediately with mechanistic explanation: \"challenging the assumption that experience protects against AI-induced error.\" Calibrate every claim to the study design, estimand, uncertainty, and strength of evidence. Primary versus secondary status determines emphasis, not whether uncertainty is acknowledged; describe subgroup and secondary analyses as exploratory or hypothesis-generating when appropriate.\n\n## Quick self-check before delivering prose in this register\n\n- Subject-verb-qualification order maintained? No throat-clearing lead-in.\n- Every \"demonstrate\u002Findicate\u002Fsuggest\u002Freveal\" matched to the actual strength of the evidence behind it?\n- Every \"significant\" tied to a prespecified test with its estimate and CI; every \"substantial\" to an effect size and interval; every \"robust\" to named sensitivity analyses?\n- \"We\" used for interpretation and authorial choices; passive reserved for apparatus-level procedure?\n- Em-dashes used only to embed definitions? Non-definitional em-dashes rewritten; density kept low even when uses are definitional?\n- Any rhetorical question immediately answered methodologically?\n- Limitations confined to their own Discussion subsection, phrased neutrally, not scattered through the narrative as asides?\n","# Clinical Research Voice\n\n## Skill Value\n\n- **Problem it solves:** Clinical AI prose can blur evidence strength, authorial responsibility, and the boundary between narrative interpretation and factual reporting.\n- **Use when:** Drafting or revising an Introduction, Discussion, narrative abstract text, cover letter, grant text, or related scientific prose.\n- **Output:** Evidence-calibrated, clinically detached prose with a focused voice self-check.\n- **Ask only if:** The manuscript section or intended scientific register cannot be inferred from the supplied text.\n- **Do not do:** Do not apply interpretive voice rules to Methods or Results, and do not invent evidence or citations.\n\nHouse style for formal scientific reporting in clinical AI research: precise, methodologically transparent, hedged only where evidence warrants, direct where findings are clear. Every claim traces to data or citation. The quoted sentences and statistics in this skill illustrate sentence form only; they are not sourced findings and must not be reused as evidence or citations.\n\n**Scope.** This voice governs narrative prose — Abstract background\u002Fconclusions, Introduction, Discussion, cover letters, grant narrative. It does not govern Methods or Results, which are audited separately under `method-section-check` and must stay purely factual and procedural. Applying this skill's interpretive latitude (hedging, mechanism, implications) to Methods or Results is itself an error — check which section you're in before applying these rules.\n\n## Sentence Architecture\n\nOpen sentences with the grammatical subject, then the verb, then qualification: \"Physicians demonstrate substantial automation bias when exposed to erroneous LLM recommendations, even with voluntary consultation and prior AI literacy training.\" Avoid front-loaded rhetorical flourishes. State the point, then elaborate — do not build toward a conclusion.\n\nVary sentence length deliberately: short declarative anchors (\"Most diagnostic errors arise from judgment-related pitfalls\") alongside complex subordinated sentences of 40+ words that specify population, condition, and effect size in a single unit.\n\nWithin a paragraph, sequence: (1) the finding or problem, (2) mechanism or context, (3) connection to implications or prior literature. A prior-literature statement carries a citation to a source that was read and resolves to a PMID or DOI; obtain it through `method-write`'s evidence access or `method-citations`, never from memory.\n\n## Pronouns and Point of View\n\nUse \"we\" for authorial actions and interpretation: \"we conducted,\" \"we find,\" \"we evaluated,\" \"we interpret this as.\" Reserve passive constructions (\"participants were randomly assigned,\" \"scoring rubrics were created\") for procedural steps where the actor is the study apparatus, not the authors — this passive register belongs primarily to Methods, not to Discussion prose.\n\nNever use \"I.\" Address readers or future researchers as implied subjects (\"Future research should explore\") rather than as \"you.\"\n\n## Punctuation Rules\n\nUse em-dashes to embed a definition mid-sentence: \"automation bias — the tendency to over-rely on automated output.\" Definitional embedding is the sanctioned use in this house style. Non-definitional em-dashes (for suspense, rhythm, or as a comma substitute) are a tell — rewrite them with commas, semicolons, or parentheticals. High em-dash density is itself a detector signal even when each use is definitional; if a paragraph accumulates several, convert all but the most load-bearing definition to parenthetical or appositive form.\n\nUse commas and semicolons to link closely related independent clauses. Avoid ellipses. Use parentheticals for statistics: \"(95% CI, −18.9 to −9.1; P\u003C.001).\" Format P values to the target journal's convention (leading zero, decimal places, and the lowest reported bound, commonly P\u003C.001 or P\u003C0.001); do not report P\u003C0.0001 unless the journal allows it.\n\nRhetorical questions are permitted only to frame an evidence gap — \"Are AI-trained physicians exercising voluntary consultation vulnerable to automation bias when LLM recommendations contain errors?\" — and must be followed immediately by the methodological response. Never leave one dangling.\n\n## Vocabulary Instructions\n\nPrefer precise technical nouns over informal approximations: \"hallucinate,\" \"automation bias,\" \"intraclass correlation coefficient,\" \"intention-to-treat,\" \"clinical vignettes.\"\n\nCalibrate verbs to strength of evidence: \"demonstrate\" for strong evidence, \"indicate\"\u002F\"reveal\" for direct findings, \"suggest\" for exploratory findings. Reserve \"significant\" for the result of a prespecified hypothesis test, reported with its estimate and confidence interval; never use it as a synonym for \"large\" or \"important.\" Use \"substantial\" only alongside an effect size and interval, and \"robust\" only when named sensitivity analyses agree (ASA statement on P values, Am Stat 2016;70:129-33, principles 3 and 5). Apply hedging phrases to subgroup and secondary analyses specifically: \"should be regarded as hypothesis-generating rather than conclusive.\"\n\nDraw analogies from clinical practice and cognitive psychology where they clarify mechanism: \"illness scripts,\" \"cognitive offloading,\" \"anchoring,\" \"fluency-induced heuristic processing.\" Analogy is for the Discussion's mechanistic explanation, not for Methods or Results.\n\n## Tone and Emotional Range\n\nMaintain clinical detachment throughout. State alarming findings plainly, without amplification: \"raising important patient safety concerns for clinical integration.\" Acknowledge limitations in a dedicated Discussion subsection using neutral concessive constructions: \"Several limitations of this study warrant consideration.\" Do not editorialize.\n\nWhen findings challenge assumptions, say so directly and follow immediately with mechanistic explanation: \"challenging the assumption that experience protects against AI-induced error.\" Calibrate every claim to the study design, estimand, uncertainty, and strength of evidence. Primary versus secondary status determines emphasis, not whether uncertainty is acknowledged; describe subgroup and secondary analyses as exploratory or hypothesis-generating when appropriate.\n\n## Quick self-check before delivering prose in this register\n\n- Subject-verb-qualification order maintained? No throat-clearing lead-in.\n- Every \"demonstrate\u002Findicate\u002Fsuggest\u002Freveal\" matched to the actual strength of the evidence behind it?\n- Every \"significant\" tied to a prespecified test with its estimate and CI; every \"substantial\" to an effect size and interval; every \"robust\" to named sensitivity analyses?\n- \"We\" used for interpretation and authorial choices; passive reserved for apparatus-level procedure?\n- Em-dashes used only to embed definitions? Non-definitional em-dashes rewritten; density kept low even when uses are definitional?\n- Any rhetorical question immediately answered methodologically?\n- Limitations confined to their own Discussion subsection, phrased neutrally, not scattered through the narrative as asides?\n",{"name":5,"description":7},"main","893bccb980e4dce1345919e9f675f5a9dcbed364","plugins\u002Fmethodsmith\u002Fskills\u002Fmethod-voice",0,"https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith","https:\u002F\u002Fsajor2000.github.io\u002Fmethodsmith\u002F","https:\u002F\u002Fgithub.com\u002Fsajor2000","Critical care physician-scientist at Rush University. @Common-Longitudinal-ICU-data-Format contributor building open ICU data tools & clinical AI","2026-11-01 01:56:01",{"source":8,"skillName":5,"commands":52,"prompt":62,"zipAvailable":63},[53,56,59],{"tool":54,"command":55},"npx","npx skills add sajor2000\u002Fmethodsmith --skill method-voice",{"tool":57,"command":58},"bunx","bunx skills add sajor2000\u002Fmethodsmith --skill method-voice",{"tool":60,"command":61},"pnpm","pnpm dlx skills add sajor2000\u002Fmethodsmith --skill method-voice","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Fplugins\u002Fmethodsmith\u002Fskills\u002Fmethod-voice\u002FSKILL.md",true,[],[66,80,94,106,115,124],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":75,"publishedAt":37},"01M71QQY0KE1184AF58EC1J5C1","skill-one","Sample skill","https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fskill-one\u002FSKILL.md",[72],{"categoryNo":73,"categoryName":74,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[76,77,78,79],"one","skill one","sample","sample skill",{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":87,"publishedAt":37},"01M71QQW3PNK9AYB5HBMAPM72J","disabled-skill","A skill with model invocation disabled","https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fdisabled-skill\u002FSKILL.md",[86],{"categoryNo":73,"categoryName":74,"taxonomyType":21,"taxonomyTypeDesc":22},[88,89,90,91,92,93],"disabled skill","a skill","with model","with","true","model",{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":101,"publishedAt":37},"01M71QQSY02X7DNZQRAGJ6YFGY","claude-only-skill","A skill restricted to Claude Code only","https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fclaude-only-skill\u002FSKILL.md",[100],{"categoryNo":73,"categoryName":74,"taxonomyType":21,"taxonomyTypeDesc":22},[102,103,104,89,105],"claude only","claude only skill","only","code",{"skillNo":107,"name":108,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":112,"publishedAt":37},"01M71QQRPZ76X623V5P7T1E3DD","default-skill","https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fcustom-paths\u002Fskills\u002Fdefault-skill\u002FSKILL.md",[111],{"categoryNo":73,"categoryName":74,"taxonomyType":21,"taxonomyTypeDesc":22},[113,114],"default","default skill",{"skillNo":116,"name":117,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":118,"categories":119,"keywords":121,"publishedAt":37},"01M71QQQG3FF2HFQHVH8VHA08R","custom-skill","https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fcustom-paths\u002Fcustom-skills\u002Fcustom-skill\u002FSKILL.md",[120],{"categoryNo":73,"categoryName":74,"taxonomyType":21,"taxonomyTypeDesc":22},[122,123],"custom skill","custom",{"skillNo":125,"name":126,"nameZh":6,"description":127,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":128,"categories":129,"keywords":130,"publishedAt":37},"01M71QQMBNC76141GHXBSMENSH","method-write","Drafts, revises, and audits evidence-bound manuscripts, sections, abstracts, and reviewer responses with claim provenance. Use when scientific prose must be written or revised (entry point).","https:\u002F\u002Fgithub.com\u002Fsajor2000\u002Fmethodsmith\u002Fblob\u002Fmain\u002Fplugins\u002Fmethodsmith\u002Fskills\u002Fmethod-write\u002FSKILL.md",[],[32,131,132,133,134,135,27,28,136,137,91,138,139,140],"method write","point","evidence","entry","claim","provenance","reviewer","scientific","prose","drafts"]