[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-minimax-seir-modeling":3,"mdc-eluubh-key":40,"related-org-minimax-seir-modeling":1219,"related-repo-minimax-seir-modeling":1422},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":29,"repoUrl":30,"updatedAt":31,"license":32,"forks":33,"topics":34,"repo":35,"sourceUrl":38,"mdContent":39},"seir-modeling","simulate disease transmission with SEIR models","当用户需要用仓室模型做传染病传播情景推演、选择 SIR\u002FSEIR\u002FSEIRS 模型结构、 理解 β\u002Fσ\u002Fγ\u002FR0 参数含义、做参数敏感性分析、估计峰值时间与峰值感染数、 解读或评审一篇仓室模型结果时使用。同义场景：SEIR 模型、SIR 模型、 仓室模型、传播动力学建模、基本再生数 R0、有效再生数、传染期\u002F潜伏期 参数化、疫情趋势情景模拟、干预效果推演、\"帮我跑一个 SEIR 看看峰值\"。",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"minimax","MiniMax","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fminimax.jpg","MiniMax-AI",[13,17,20,23,26],{"name":14,"slug":15,"type":16},"Research","research","tag",{"name":18,"slug":19,"type":16},"Healthcare","healthcare",{"name":21,"slug":22,"type":16},"Data Analysis","data-analysis",{"name":24,"slug":25,"type":16},"Simulation","simulation",{"name":27,"slug":28,"type":16},"Statistics","statistics",5,"https:\u002F\u002Fgithub.com\u002FMiniMax-AI\u002FMiniMax-Code-Plugins","2026-08-21T03:59:48.247344",null,2,[],{"repoUrl":30,"stars":29,"forks":33,"topics":36,"description":37},[],"Community registry and contribution toolkit for MiniMax Code plugins.","https:\u002F\u002Fgithub.com\u002FMiniMax-AI\u002FMiniMax-Code-Plugins\u002Ftree\u002FHEAD\u002Fplugins\u002FHylouis233\u002Fopenscience\u002Fskills\u002Fseir-modeling","---\nname: seir-modeling\ndescription: >-\n  当用户需要用仓室模型做传染病传播情景推演、选择 SIR\u002FSEIR\u002FSEIRS 模型结构、\n  理解 β\u002Fσ\u002Fγ\u002FR0 参数含义、做参数敏感性分析、估计峰值时间与峰值感染数、\n  解读或评审一篇仓室模型结果时使用。同义场景：SEIR 模型、SIR 模型、\n  仓室模型、传播动力学建模、基本再生数 R0、有效再生数、传染期\u002F潜伏期\n  参数化、疫情趋势情景模拟、干预效果推演、\"帮我跑一个 SEIR 看看峰值\"。\nargument-hint: '[model=sir|seir beta sigma gamma population i0 days]'\nmetadata:\n  domains: [epidemiology, public-health]\n  last_reviewed: '2026-08-18'\n---\n\n# seir-modeling：仓室模型情景推演规程\n\n## 目的\n\n用确定性仓室模型（SIR\u002FSEIR）做**情景推演**：在给定参数假设下，回答\n\"若传播机制如此、参数取这些值，曲线大致长什么样、峰值何时多高\"。\n本技能规定模型选择的决策树、参数意义与来源标注、敏感性分析做法，\n以及最重要的**解读纪律**——模型输出是条件推演，不是预测水晶球。\n\n解读纪律（内化后再动手）：\n\n1. 每个数字都连着假设：报告结果时必须同段给出参数取值、取值来源与\n   区间；光秃秃的峰值数字不得写入产物。\n2. 模型是简化：均匀混合、无年龄结构、无空间、参数恒定。现实偏离\n   这些假设时，输出只作定性参考。\n3. 参数未知时跑**区间**而非单点：用多组参数给出情景范围，杜绝\n   \"挑一个看起来合理的数跑一次就当结论\"。\n\n## 前置检查\n\n1. 明确推演问题：要回答\"峰值多高\"\"何时达峰\"还是\"某参数变化对曲线\n   形态的影响\"？问题决定模型与输出重点。\n2. 确认病种关键生物学参数（潜伏期、传染期）有文献或方案来源；拿不到\n   时标 `[模型知识—待核实]` 并在报告中说明这是假设值而非实测值。\n3. 确认人群规模与初始感染数的设定依据（真实场景规模 or 示意性规模）。\n4. 脚本就绪：`skills\u002Fseir-modeling\u002Fscripts\u002Fseir.py`（纯 stdlib）。\n\n## 1 · 模型选择决策树\n\n```\n是否需要显式表达\"已感染但尚未有传染性\"的潜伏期？\n├─ 不需要（潜伏期很短或只关心粗略形态）→ SIR\n└─ 需要（潜伏期不可忽视）→ SEIR\n     └─ 康复后免疫是否随时间消失？\n          ├─ 不消失（本脚本范围）→ SEIR\n          └─ 消失（需 R → S 回流）→ SEIRS（本脚本不实现，见\"不做什么\"）\n```\n\n补充判断：\n\n- 只关心**单波封闭人群**（如一次集体单位暴发后的续发），SIR\u002FSEIR 够用；\n- 需要出生死亡、年龄结构、空间或随机效应，超出本脚本范围，转\n  计算类 skill 或专业建模环境；\n- 不确定选哪个：两个都跑，比较形态差异，把差异写进报告（这本身\n  就是敏感性分析）。\n\n## 2 · 参数意义与标定\n\n| 参数 | 含义 | 与可观测量的关系 |\n| --- | --- | --- |\n| β（beta） | 传播率：单位时间有效接触 × 单次接触传播概率 | 不直接可观测，常由 R0 反推 |\n| σ（sigma） | 潜伏期转化率 | ≈ 1\u002F平均潜伏期 [模型知识—待核实：指数分布假设下成立] |\n| γ（gamma） | 恢复率 | ≈ 1\u002F平均传染期 [模型知识—待核实：同上] |\n| R0 | 基本再生数 | SIR 及本脚本 SEIR 形式下 R0 = β\u002Fγ [模型知识—待核实] |\n\n标定规程：\n\n1. 先定 γ 与 σ（来自潜伏期\u002F传染期的文献或方案值，带来源标签）；\n2. 再由目标 R0 反推 β = R0·γ，R0 的取值来源同样标注；\n3. 初始值：i0（初始感染者）、e0（初始潜伏者，默认 0）、population；\n4. 每个参数在报告中给\"取值 + 来源标签 + 不确定区间\"三件套，缺一\n   不写结果。\n\n## 3 · 运行脚本\n\n```bash\n# JSON 输出（含参数回显、R0、峰值、逐日序列）\npython scripts\u002Fseir.py --model seir --beta 0.5 --sigma 0.2 --gamma 0.1 \\\n    --population 1000000 --i0 10 --days 160 --format json\n\n# CSV 输出（序列到 stdout，峰值摘要到 stderr）\npython scripts\u002Fseir.py --model sir --beta 0.3 --gamma 0.1 \\\n    --population 100000 --i0 5 --days 120 --format csv > sir.csv\n```\n\n脚本行为说明：\n\n- 积分器为手写 RK4，步长 `--dt`（默认 0.1 天），逐日快照输出；\n- 输入校验：参数必须为正，i0+e0 ≤ population，SEIR 必须给 σ；\n- 输出中的 `r0` 由 β\u002Fγ 计算回显，供核对参数录入是否有误。\n\n## 4 · 敏感性分析\n\n至少做两组对比，全部落盘并登记 provenance：\n\n1. **参数区间扫描**：对每个不确定参数取其区间端点（如 R0 的文献\n   上下限），组合出低\u002F中\u002F高三档情景各跑一次；\n2. **结构敏感性**：SIR 与 SEIR 各跑一次，比较峰值时间与高度差异。\n\n报告形式：情景表（每行一个情景：参数取值、峰值时间、峰值感染数、\n最终规模），不给\"最优拟合\"单一结论。需要系统拟合观测数据（最小\n二乘\u002FMCMC）时转计算类 skill，本技能不拟合。\n\n## 5 · 结果解读纪律\n\n写入报告时逐条自查：\n\n1. 每个数字旁边是否有参数三件套（取值\u002F来源\u002F区间）？\n2. 措辞是否为\"在……假设下，推演显示……\"，而非\"预测将……\"？\n3. 是否说明了模型简化（均匀混合、参数恒定、封闭人群）及偏离方向？\n4. 峰值是否同时给出**区间**（多档情景的峰值的范围）？\n5. 若把推演与真实监测曲线并置，是否明确标注\"模型未拟合该数据\"\n   或拟合方法？\n6. 任何对外（含媒体、决策部门）使用推演结果，是否已按 guardrail\n   第 8 条取得确认，并附完整假设说明？\n\n## 输出模板\n\n```markdown\n## 仓室模型情景推演（\u003C日期>）\n\n### 模型与假设\n- 结构：SEIR；均匀混合、封闭人群、参数恒定（简化假设）\n- 参数：β=\u003C值>（由 R0=\u003C值>×γ 反推，R0 来源 \u003C标签>）；\n  σ=\u003C值>（≈1\u002F潜伏期 \u003C值> 天，\u003C来源标签>）；γ=\u003C值>（≈1\u002F传染期，\u003C来源标签>）\n- 初始：N=\u003Cpopulation>，I₀=\u003Ci0>，E₀=\u003Ce0>\n\n### 情景结果\n| 情景 | R0 | 峰日 | 峰值感染数 | 备注 |\n| --- | --- | --- | --- | --- |\n| 低 | … | … | … | [实验数据：seir.py 输出] |\n\n### 解读\n- 在以上假设下，推演显示 \u003C峰值区间与达峰时间区间>；这些是条件推演，\n  不构成疫情预测。\n- 敏感性：\u003C哪个参数最影响结果>；结构对比：\u003CSIR vs SEIR 差异>。\n\n### 局限\n- \u003C逐条列出简化假设与数据缺口>\n```\n\n## 本技能不做什么\n\n- **不做参数拟合**：不用观测数据估计参数（最小二乘\u002F极大似然\u002FMCMC\n  交给计算类 skill 或专业环境）；本技能只做前向推演。\n- **不做 SEIRS\u002FSEIRV\u002F年龄结构\u002F随机模型**：脚本只实现 SIR 与 SEIR；\n  需要更复杂结构时说明理由并转专业工具。\n- **不做实时预测与对外发布**：推演结果对外使用属 guardrail 第 8 条\n  危险操作，必须先确认；任何\"疫情预测\"措辞在产物中禁用。\n- **不编造病种参数**：潜伏期\u002F传染期\u002FR0 的具体数值必须有来源；\n  模型印象值标 `[模型知识—待核实]` 且只用于演示性情景。\n- **不做政策建议**：干预参数（如β下降多少对应何种措施）的映射超出\n  本技能证据范围，不替用户把情景翻译成决策。\n\n## 收尾与下一步\n\n1. 汇总：跑了几个情景、峰值与达峰时间的区间、最敏感的参数。\n2. 指向产物（JSON\u002FCSV 序列、情景报告）并登记 provenance。\n3. 建议下一步：\n   - 需要拟合真实数据或更复杂模型 → python-analysis 等计算类 skill；\n   - 结果写入报告 → `epi-writing`（附参数三件套与情景表）；\n   - 参数来源需补文献 → literature-search。\n",{"data":41,"body":48},{"name":4,"description":6,"argument-hint":42,"metadata":43},"[model=sir|seir beta sigma gamma population i0 days]",{"domains":44,"last_reviewed":47},[45,46],"epidemiology","public-health","2026-08-18",{"type":49,"children":50},"root",[51,60,66,87,92,119,124,164,170,182,187,213,219,340,345,368,374,611,616,650,656,661,684,689,695,700,739,744,1099,1104,1164,1169,1213],{"type":52,"tag":53,"props":54,"children":56},"element","h1",{"id":55},"seir-modeling仓室模型情景推演规程",[57],{"type":58,"value":59},"text","seir-modeling：仓室模型情景推演规程",{"type":52,"tag":61,"props":62,"children":64},"h2",{"id":63},"目的",[65],{"type":58,"value":63},{"type":52,"tag":67,"props":68,"children":69},"p",{},[70,72,78,80,85],{"type":58,"value":71},"用确定性仓室模型（SIR\u002FSEIR）做",{"type":52,"tag":73,"props":74,"children":75},"strong",{},[76],{"type":58,"value":77},"情景推演",{"type":58,"value":79},"：在给定参数假设下，回答\n\"若传播机制如此、参数取这些值，曲线大致长什么样、峰值何时多高\"。\n本技能规定模型选择的决策树、参数意义与来源标注、敏感性分析做法，\n以及最重要的",{"type":52,"tag":73,"props":81,"children":82},{},[83],{"type":58,"value":84},"解读纪律",{"type":58,"value":86},"——模型输出是条件推演，不是预测水晶球。",{"type":52,"tag":67,"props":88,"children":89},{},[90],{"type":58,"value":91},"解读纪律（内化后再动手）：",{"type":52,"tag":93,"props":94,"children":95},"ol",{},[96,102,107],{"type":52,"tag":97,"props":98,"children":99},"li",{},[100],{"type":58,"value":101},"每个数字都连着假设：报告结果时必须同段给出参数取值、取值来源与\n区间；光秃秃的峰值数字不得写入产物。",{"type":52,"tag":97,"props":103,"children":104},{},[105],{"type":58,"value":106},"模型是简化：均匀混合、无年龄结构、无空间、参数恒定。现实偏离\n这些假设时，输出只作定性参考。",{"type":52,"tag":97,"props":108,"children":109},{},[110,112,117],{"type":58,"value":111},"参数未知时跑",{"type":52,"tag":73,"props":113,"children":114},{},[115],{"type":58,"value":116},"区间",{"type":58,"value":118},"而非单点：用多组参数给出情景范围，杜绝\n\"挑一个看起来合理的数跑一次就当结论\"。",{"type":52,"tag":61,"props":120,"children":122},{"id":121},"前置检查",[123],{"type":58,"value":121},{"type":52,"tag":93,"props":125,"children":126},{},[127,132,146,151],{"type":52,"tag":97,"props":128,"children":129},{},[130],{"type":58,"value":131},"明确推演问题：要回答\"峰值多高\"\"何时达峰\"还是\"某参数变化对曲线\n形态的影响\"？问题决定模型与输出重点。",{"type":52,"tag":97,"props":133,"children":134},{},[135,137,144],{"type":58,"value":136},"确认病种关键生物学参数（潜伏期、传染期）有文献或方案来源；拿不到\n时标 ",{"type":52,"tag":138,"props":139,"children":141},"code",{"className":140},[],[142],{"type":58,"value":143},"[模型知识—待核实]",{"type":58,"value":145}," 并在报告中说明这是假设值而非实测值。",{"type":52,"tag":97,"props":147,"children":148},{},[149],{"type":58,"value":150},"确认人群规模与初始感染数的设定依据（真实场景规模 or 示意性规模）。",{"type":52,"tag":97,"props":152,"children":153},{},[154,156,162],{"type":58,"value":155},"脚本就绪：",{"type":52,"tag":138,"props":157,"children":159},{"className":158},[],[160],{"type":58,"value":161},"skills\u002Fseir-modeling\u002Fscripts\u002Fseir.py",{"type":58,"value":163},"（纯 stdlib）。",{"type":52,"tag":61,"props":165,"children":167},{"id":166},"_1-模型选择决策树",[168],{"type":58,"value":169},"1 · 模型选择决策树",{"type":52,"tag":171,"props":172,"children":176},"pre",{"className":173,"code":175,"language":58},[174],"language-text","是否需要显式表达\"已感染但尚未有传染性\"的潜伏期？\n├─ 不需要（潜伏期很短或只关心粗略形态）→ SIR\n└─ 需要（潜伏期不可忽视）→ SEIR\n     └─ 康复后免疫是否随时间消失？\n          ├─ 不消失（本脚本范围）→ SEIR\n          └─ 消失（需 R → S 回流）→ SEIRS（本脚本不实现，见\"不做什么\"）\n",[177],{"type":52,"tag":138,"props":178,"children":180},{"__ignoreMap":179},"",[181],{"type":58,"value":175},{"type":52,"tag":67,"props":183,"children":184},{},[185],{"type":58,"value":186},"补充判断：",{"type":52,"tag":188,"props":189,"children":190},"ul",{},[191,203,208],{"type":52,"tag":97,"props":192,"children":193},{},[194,196,201],{"type":58,"value":195},"只关心",{"type":52,"tag":73,"props":197,"children":198},{},[199],{"type":58,"value":200},"单波封闭人群",{"type":58,"value":202},"（如一次集体单位暴发后的续发），SIR\u002FSEIR 够用；",{"type":52,"tag":97,"props":204,"children":205},{},[206],{"type":58,"value":207},"需要出生死亡、年龄结构、空间或随机效应，超出本脚本范围，转\n计算类 skill 或专业建模环境；",{"type":52,"tag":97,"props":209,"children":210},{},[211],{"type":58,"value":212},"不确定选哪个：两个都跑，比较形态差异，把差异写进报告（这本身\n就是敏感性分析）。",{"type":52,"tag":61,"props":214,"children":216},{"id":215},"_2-参数意义与标定",[217],{"type":58,"value":218},"2 · 参数意义与标定",{"type":52,"tag":220,"props":221,"children":222},"table",{},[223,247],{"type":52,"tag":224,"props":225,"children":226},"thead",{},[227],{"type":52,"tag":228,"props":229,"children":230},"tr",{},[231,237,242],{"type":52,"tag":232,"props":233,"children":234},"th",{},[235],{"type":58,"value":236},"参数",{"type":52,"tag":232,"props":238,"children":239},{},[240],{"type":58,"value":241},"含义",{"type":52,"tag":232,"props":243,"children":244},{},[245],{"type":58,"value":246},"与可观测量的关系",{"type":52,"tag":248,"props":249,"children":250},"tbody",{},[251,270,294,317],{"type":52,"tag":228,"props":252,"children":253},{},[254,260,265],{"type":52,"tag":255,"props":256,"children":257},"td",{},[258],{"type":58,"value":259},"β（beta）",{"type":52,"tag":255,"props":261,"children":262},{},[263],{"type":58,"value":264},"传播率：单位时间有效接触 × 单次接触传播概率",{"type":52,"tag":255,"props":266,"children":267},{},[268],{"type":58,"value":269},"不直接可观测，常由 R0 反推",{"type":52,"tag":228,"props":271,"children":272},{},[273,278,283],{"type":52,"tag":255,"props":274,"children":275},{},[276],{"type":58,"value":277},"σ（sigma）",{"type":52,"tag":255,"props":279,"children":280},{},[281],{"type":58,"value":282},"潜伏期转化率",{"type":52,"tag":255,"props":284,"children":285},{},[286,288],{"type":58,"value":287},"≈ 1\u002F平均潜伏期 ",{"type":52,"tag":289,"props":290,"children":291},"span",{},[292],{"type":58,"value":293},"模型知识—待核实：指数分布假设下成立",{"type":52,"tag":228,"props":295,"children":296},{},[297,302,307],{"type":52,"tag":255,"props":298,"children":299},{},[300],{"type":58,"value":301},"γ（gamma）",{"type":52,"tag":255,"props":303,"children":304},{},[305],{"type":58,"value":306},"恢复率",{"type":52,"tag":255,"props":308,"children":309},{},[310,312],{"type":58,"value":311},"≈ 1\u002F平均传染期 ",{"type":52,"tag":289,"props":313,"children":314},{},[315],{"type":58,"value":316},"模型知识—待核实：同上",{"type":52,"tag":228,"props":318,"children":319},{},[320,325,330],{"type":52,"tag":255,"props":321,"children":322},{},[323],{"type":58,"value":324},"R0",{"type":52,"tag":255,"props":326,"children":327},{},[328],{"type":58,"value":329},"基本再生数",{"type":52,"tag":255,"props":331,"children":332},{},[333,335],{"type":58,"value":334},"SIR 及本脚本 SEIR 形式下 R0 = β\u002Fγ ",{"type":52,"tag":289,"props":336,"children":337},{},[338],{"type":58,"value":339},"模型知识—待核实",{"type":52,"tag":67,"props":341,"children":342},{},[343],{"type":58,"value":344},"标定规程：",{"type":52,"tag":93,"props":346,"children":347},{},[348,353,358,363],{"type":52,"tag":97,"props":349,"children":350},{},[351],{"type":58,"value":352},"先定 γ 与 σ（来自潜伏期\u002F传染期的文献或方案值，带来源标签）；",{"type":52,"tag":97,"props":354,"children":355},{},[356],{"type":58,"value":357},"再由目标 R0 反推 β = R0·γ，R0 的取值来源同样标注；",{"type":52,"tag":97,"props":359,"children":360},{},[361],{"type":58,"value":362},"初始值：i0（初始感染者）、e0（初始潜伏者，默认 0）、population；",{"type":52,"tag":97,"props":364,"children":365},{},[366],{"type":58,"value":367},"每个参数在报告中给\"取值 + 来源标签 + 不确定区间\"三件套，缺一\n不写结果。",{"type":52,"tag":61,"props":369,"children":371},{"id":370},"_3-运行脚本",[372],{"type":58,"value":373},"3 · 运行脚本",{"type":52,"tag":171,"props":375,"children":379},{"className":376,"code":377,"language":378,"meta":179,"style":179},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","# JSON 输出（含参数回显、R0、峰值、逐日序列）\npython scripts\u002Fseir.py --model seir --beta 0.5 --sigma 0.2 --gamma 0.1 \\\n    --population 1000000 --i0 10 --days 160 --format json\n\n# CSV 输出（序列到 stdout，峰值摘要到 stderr）\npython scripts\u002Fseir.py --model sir --beta 0.3 --gamma 0.1 \\\n    --population 100000 --i0 5 --days 120 --format csv > sir.csv\n","bash",[380],{"type":52,"tag":138,"props":381,"children":382},{"__ignoreMap":179},[383,394,456,500,510,518,560],{"type":52,"tag":289,"props":384,"children":387},{"class":385,"line":386},"line",1,[388],{"type":52,"tag":289,"props":389,"children":391},{"style":390},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[392],{"type":58,"value":393},"# JSON 输出（含参数回显、R0、峰值、逐日序列）\n",{"type":52,"tag":289,"props":395,"children":396},{"class":385,"line":33},[397,403,409,414,419,424,430,435,440,445,450],{"type":52,"tag":289,"props":398,"children":400},{"style":399},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[401],{"type":58,"value":402},"python",{"type":52,"tag":289,"props":404,"children":406},{"style":405},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[407],{"type":58,"value":408}," scripts\u002Fseir.py",{"type":52,"tag":289,"props":410,"children":411},{"style":405},[412],{"type":58,"value":413}," --model",{"type":52,"tag":289,"props":415,"children":416},{"style":405},[417],{"type":58,"value":418}," seir",{"type":52,"tag":289,"props":420,"children":421},{"style":405},[422],{"type":58,"value":423}," --beta",{"type":52,"tag":289,"props":425,"children":427},{"style":426},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[428],{"type":58,"value":429}," 0.5",{"type":52,"tag":289,"props":431,"children":432},{"style":405},[433],{"type":58,"value":434}," --sigma",{"type":52,"tag":289,"props":436,"children":437},{"style":426},[438],{"type":58,"value":439}," 0.2",{"type":52,"tag":289,"props":441,"children":442},{"style":405},[443],{"type":58,"value":444}," --gamma",{"type":52,"tag":289,"props":446,"children":447},{"style":426},[448],{"type":58,"value":449}," 0.1",{"type":52,"tag":289,"props":451,"children":453},{"style":452},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[454],{"type":58,"value":455}," \\\n",{"type":52,"tag":289,"props":457,"children":459},{"class":385,"line":458},3,[460,465,470,475,480,485,490,495],{"type":52,"tag":289,"props":461,"children":462},{"style":405},[463],{"type":58,"value":464},"    --population",{"type":52,"tag":289,"props":466,"children":467},{"style":426},[468],{"type":58,"value":469}," 1000000",{"type":52,"tag":289,"props":471,"children":472},{"style":405},[473],{"type":58,"value":474}," --i0",{"type":52,"tag":289,"props":476,"children":477},{"style":426},[478],{"type":58,"value":479}," 10",{"type":52,"tag":289,"props":481,"children":482},{"style":405},[483],{"type":58,"value":484}," --days",{"type":52,"tag":289,"props":486,"children":487},{"style":426},[488],{"type":58,"value":489}," 160",{"type":52,"tag":289,"props":491,"children":492},{"style":405},[493],{"type":58,"value":494}," --format",{"type":52,"tag":289,"props":496,"children":497},{"style":405},[498],{"type":58,"value":499}," json\n",{"type":52,"tag":289,"props":501,"children":503},{"class":385,"line":502},4,[504],{"type":52,"tag":289,"props":505,"children":507},{"emptyLinePlaceholder":506},true,[508],{"type":58,"value":509},"\n",{"type":52,"tag":289,"props":511,"children":512},{"class":385,"line":29},[513],{"type":52,"tag":289,"props":514,"children":515},{"style":390},[516],{"type":58,"value":517},"# CSV 输出（序列到 stdout，峰值摘要到 stderr）\n",{"type":52,"tag":289,"props":519,"children":521},{"class":385,"line":520},6,[522,526,530,534,539,543,548,552,556],{"type":52,"tag":289,"props":523,"children":524},{"style":399},[525],{"type":58,"value":402},{"type":52,"tag":289,"props":527,"children":528},{"style":405},[529],{"type":58,"value":408},{"type":52,"tag":289,"props":531,"children":532},{"style":405},[533],{"type":58,"value":413},{"type":52,"tag":289,"props":535,"children":536},{"style":405},[537],{"type":58,"value":538}," sir",{"type":52,"tag":289,"props":540,"children":541},{"style":405},[542],{"type":58,"value":423},{"type":52,"tag":289,"props":544,"children":545},{"style":426},[546],{"type":58,"value":547}," 0.3",{"type":52,"tag":289,"props":549,"children":550},{"style":405},[551],{"type":58,"value":444},{"type":52,"tag":289,"props":553,"children":554},{"style":426},[555],{"type":58,"value":449},{"type":52,"tag":289,"props":557,"children":558},{"style":452},[559],{"type":58,"value":455},{"type":52,"tag":289,"props":561,"children":563},{"class":385,"line":562},7,[564,568,573,577,582,586,591,595,600,606],{"type":52,"tag":289,"props":565,"children":566},{"style":405},[567],{"type":58,"value":464},{"type":52,"tag":289,"props":569,"children":570},{"style":426},[571],{"type":58,"value":572}," 100000",{"type":52,"tag":289,"props":574,"children":575},{"style":405},[576],{"type":58,"value":474},{"type":52,"tag":289,"props":578,"children":579},{"style":426},[580],{"type":58,"value":581}," 5",{"type":52,"tag":289,"props":583,"children":584},{"style":405},[585],{"type":58,"value":484},{"type":52,"tag":289,"props":587,"children":588},{"style":426},[589],{"type":58,"value":590}," 120",{"type":52,"tag":289,"props":592,"children":593},{"style":405},[594],{"type":58,"value":494},{"type":52,"tag":289,"props":596,"children":597},{"style":405},[598],{"type":58,"value":599}," csv",{"type":52,"tag":289,"props":601,"children":603},{"style":602},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[604],{"type":58,"value":605}," >",{"type":52,"tag":289,"props":607,"children":608},{"style":405},[609],{"type":58,"value":610}," sir.csv\n",{"type":52,"tag":67,"props":612,"children":613},{},[614],{"type":58,"value":615},"脚本行为说明：",{"type":52,"tag":188,"props":617,"children":618},{},[619,632,637],{"type":52,"tag":97,"props":620,"children":621},{},[622,624,630],{"type":58,"value":623},"积分器为手写 RK4，步长 ",{"type":52,"tag":138,"props":625,"children":627},{"className":626},[],[628],{"type":58,"value":629},"--dt",{"type":58,"value":631},"（默认 0.1 天），逐日快照输出；",{"type":52,"tag":97,"props":633,"children":634},{},[635],{"type":58,"value":636},"输入校验：参数必须为正，i0+e0 ≤ population，SEIR 必须给 σ；",{"type":52,"tag":97,"props":638,"children":639},{},[640,642,648],{"type":58,"value":641},"输出中的 ",{"type":52,"tag":138,"props":643,"children":645},{"className":644},[],[646],{"type":58,"value":647},"r0",{"type":58,"value":649}," 由 β\u002Fγ 计算回显，供核对参数录入是否有误。",{"type":52,"tag":61,"props":651,"children":653},{"id":652},"_4-敏感性分析",[654],{"type":58,"value":655},"4 · 敏感性分析",{"type":52,"tag":67,"props":657,"children":658},{},[659],{"type":58,"value":660},"至少做两组对比，全部落盘并登记 provenance：",{"type":52,"tag":93,"props":662,"children":663},{},[664,674],{"type":52,"tag":97,"props":665,"children":666},{},[667,672],{"type":52,"tag":73,"props":668,"children":669},{},[670],{"type":58,"value":671},"参数区间扫描",{"type":58,"value":673},"：对每个不确定参数取其区间端点（如 R0 的文献\n上下限），组合出低\u002F中\u002F高三档情景各跑一次；",{"type":52,"tag":97,"props":675,"children":676},{},[677,682],{"type":52,"tag":73,"props":678,"children":679},{},[680],{"type":58,"value":681},"结构敏感性",{"type":58,"value":683},"：SIR 与 SEIR 各跑一次，比较峰值时间与高度差异。",{"type":52,"tag":67,"props":685,"children":686},{},[687],{"type":58,"value":688},"报告形式：情景表（每行一个情景：参数取值、峰值时间、峰值感染数、\n最终规模），不给\"最优拟合\"单一结论。需要系统拟合观测数据（最小\n二乘\u002FMCMC）时转计算类 skill，本技能不拟合。",{"type":52,"tag":61,"props":690,"children":692},{"id":691},"_5-结果解读纪律",[693],{"type":58,"value":694},"5 · 结果解读纪律",{"type":52,"tag":67,"props":696,"children":697},{},[698],{"type":58,"value":699},"写入报告时逐条自查：",{"type":52,"tag":93,"props":701,"children":702},{},[703,708,713,718,729,734],{"type":52,"tag":97,"props":704,"children":705},{},[706],{"type":58,"value":707},"每个数字旁边是否有参数三件套（取值\u002F来源\u002F区间）？",{"type":52,"tag":97,"props":709,"children":710},{},[711],{"type":58,"value":712},"措辞是否为\"在……假设下，推演显示……\"，而非\"预测将……\"？",{"type":52,"tag":97,"props":714,"children":715},{},[716],{"type":58,"value":717},"是否说明了模型简化（均匀混合、参数恒定、封闭人群）及偏离方向？",{"type":52,"tag":97,"props":719,"children":720},{},[721,723,727],{"type":58,"value":722},"峰值是否同时给出",{"type":52,"tag":73,"props":724,"children":725},{},[726],{"type":58,"value":116},{"type":58,"value":728},"（多档情景的峰值的范围）？",{"type":52,"tag":97,"props":730,"children":731},{},[732],{"type":58,"value":733},"若把推演与真实监测曲线并置，是否明确标注\"模型未拟合该数据\"\n或拟合方法？",{"type":52,"tag":97,"props":735,"children":736},{},[737],{"type":58,"value":738},"任何对外（含媒体、决策部门）使用推演结果，是否已按 guardrail\n第 8 条取得确认，并附完整假设说明？",{"type":52,"tag":61,"props":740,"children":742},{"id":741},"输出模板",[743],{"type":58,"value":741},{"type":52,"tag":171,"props":745,"children":749},{"className":746,"code":747,"language":748,"meta":179,"style":179},"language-markdown shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","## 仓室模型情景推演（\u003C日期>）\n\n### 模型与假设\n- 结构：SEIR；均匀混合、封闭人群、参数恒定（简化假设）\n- 参数：β=\u003C值>（由 R0=\u003C值>×γ 反推，R0 来源 \u003C标签>）；\n  σ=\u003C值>（≈1\u002F潜伏期 \u003C值> 天，\u003C来源标签>）；γ=\u003C值>（≈1\u002F传染期，\u003C来源标签>）\n- 初始：N=\u003Cpopulation>，I₀=\u003Ci0>，E₀=\u003Ce0>\n\n### 情景结果\n| 情景 | R0 | 峰日 | 峰值感染数 | 备注 |\n| --- | --- | --- | --- | --- |\n| 低 | … | … | … | [实验数据：seir.py 输出] |\n\n### 解读\n- 在以上假设下，推演显示 \u003C峰值区间与达峰时间区间>；这些是条件推演，\n  不构成疫情预测。\n- 敏感性：\u003C哪个参数最影响结果>；结构对比：\u003CSIR vs SEIR 差异>。\n\n### 局限\n- \u003C逐条列出简化假设与数据缺口>\n","markdown",[750],{"type":52,"tag":138,"props":751,"children":752},{"__ignoreMap":179},[753,766,773,786,799,811,819,831,839,852,907,958,1009,1017,1030,1043,1052,1065,1073,1086],{"type":52,"tag":289,"props":754,"children":755},{"class":385,"line":386},[756,761],{"type":52,"tag":289,"props":757,"children":758},{"style":602},[759],{"type":58,"value":760},"## ",{"type":52,"tag":289,"props":762,"children":763},{"style":399},[764],{"type":58,"value":765},"仓室模型情景推演（\u003C日期>）\n",{"type":52,"tag":289,"props":767,"children":768},{"class":385,"line":33},[769],{"type":52,"tag":289,"props":770,"children":771},{"emptyLinePlaceholder":506},[772],{"type":58,"value":509},{"type":52,"tag":289,"props":774,"children":775},{"class":385,"line":458},[776,781],{"type":52,"tag":289,"props":777,"children":778},{"style":602},[779],{"type":58,"value":780},"### ",{"type":52,"tag":289,"props":782,"children":783},{"style":399},[784],{"type":58,"value":785},"模型与假设\n",{"type":52,"tag":289,"props":787,"children":788},{"class":385,"line":502},[789,794],{"type":52,"tag":289,"props":790,"children":791},{"style":602},[792],{"type":58,"value":793},"-",{"type":52,"tag":289,"props":795,"children":796},{"style":452},[797],{"type":58,"value":798}," 结构：SEIR；均匀混合、封闭人群、参数恒定（简化假设）\n",{"type":52,"tag":289,"props":800,"children":801},{"class":385,"line":29},[802,806],{"type":52,"tag":289,"props":803,"children":804},{"style":602},[805],{"type":58,"value":793},{"type":52,"tag":289,"props":807,"children":808},{"style":452},[809],{"type":58,"value":810}," 参数：β=\u003C值>（由 R0=\u003C值>×γ 反推，R0 来源 \u003C标签>）；\n",{"type":52,"tag":289,"props":812,"children":813},{"class":385,"line":520},[814],{"type":52,"tag":289,"props":815,"children":816},{"style":452},[817],{"type":58,"value":818},"  σ=\u003C值>（≈1\u002F潜伏期 \u003C值> 天，\u003C来源标签>）；γ=\u003C值>（≈1\u002F传染期，\u003C来源标签>）\n",{"type":52,"tag":289,"props":820,"children":821},{"class":385,"line":562},[822,826],{"type":52,"tag":289,"props":823,"children":824},{"style":602},[825],{"type":58,"value":793},{"type":52,"tag":289,"props":827,"children":828},{"style":452},[829],{"type":58,"value":830}," 初始：N=\u003Cpopulation>，I₀=\u003Ci0>，E₀=\u003Ce0>\n",{"type":52,"tag":289,"props":832,"children":834},{"class":385,"line":833},8,[835],{"type":52,"tag":289,"props":836,"children":837},{"emptyLinePlaceholder":506},[838],{"type":58,"value":509},{"type":52,"tag":289,"props":840,"children":842},{"class":385,"line":841},9,[843,847],{"type":52,"tag":289,"props":844,"children":845},{"style":602},[846],{"type":58,"value":780},{"type":52,"tag":289,"props":848,"children":849},{"style":399},[850],{"type":58,"value":851},"情景结果\n",{"type":52,"tag":289,"props":853,"children":855},{"class":385,"line":854},10,[856,861,866,870,875,879,884,888,893,897,902],{"type":52,"tag":289,"props":857,"children":858},{"style":602},[859],{"type":58,"value":860},"|",{"type":52,"tag":289,"props":862,"children":863},{"style":452},[864],{"type":58,"value":865}," 情景 ",{"type":52,"tag":289,"props":867,"children":868},{"style":602},[869],{"type":58,"value":860},{"type":52,"tag":289,"props":871,"children":872},{"style":452},[873],{"type":58,"value":874}," R0 ",{"type":52,"tag":289,"props":876,"children":877},{"style":602},[878],{"type":58,"value":860},{"type":52,"tag":289,"props":880,"children":881},{"style":452},[882],{"type":58,"value":883}," 峰日 ",{"type":52,"tag":289,"props":885,"children":886},{"style":602},[887],{"type":58,"value":860},{"type":52,"tag":289,"props":889,"children":890},{"style":452},[891],{"type":58,"value":892}," 峰值感染数 ",{"type":52,"tag":289,"props":894,"children":895},{"style":602},[896],{"type":58,"value":860},{"type":52,"tag":289,"props":898,"children":899},{"style":452},[900],{"type":58,"value":901}," 备注 ",{"type":52,"tag":289,"props":903,"children":904},{"style":602},[905],{"type":58,"value":906},"|\n",{"type":52,"tag":289,"props":908,"children":910},{"class":385,"line":909},11,[911,915,920,925,929,933,937,941,945,949,953],{"type":52,"tag":289,"props":912,"children":913},{"style":602},[914],{"type":58,"value":860},{"type":52,"tag":289,"props":916,"children":917},{"style":602},[918],{"type":58,"value":919}," ---",{"type":52,"tag":289,"props":921,"children":922},{"style":602},[923],{"type":58,"value":924}," |",{"type":52,"tag":289,"props":926,"children":927},{"style":602},[928],{"type":58,"value":919},{"type":52,"tag":289,"props":930,"children":931},{"style":602},[932],{"type":58,"value":924},{"type":52,"tag":289,"props":934,"children":935},{"style":602},[936],{"type":58,"value":919},{"type":52,"tag":289,"props":938,"children":939},{"style":602},[940],{"type":58,"value":924},{"type":52,"tag":289,"props":942,"children":943},{"style":602},[944],{"type":58,"value":919},{"type":52,"tag":289,"props":946,"children":947},{"style":602},[948],{"type":58,"value":924},{"type":52,"tag":289,"props":950,"children":951},{"style":602},[952],{"type":58,"value":919},{"type":52,"tag":289,"props":954,"children":955},{"style":602},[956],{"type":58,"value":957}," |\n",{"type":52,"tag":289,"props":959,"children":961},{"class":385,"line":960},12,[962,966,971,975,980,984,988,992,996,1000,1005],{"type":52,"tag":289,"props":963,"children":964},{"style":602},[965],{"type":58,"value":860},{"type":52,"tag":289,"props":967,"children":968},{"style":452},[969],{"type":58,"value":970}," 低 ",{"type":52,"tag":289,"props":972,"children":973},{"style":602},[974],{"type":58,"value":860},{"type":52,"tag":289,"props":976,"children":977},{"style":452},[978],{"type":58,"value":979}," … ",{"type":52,"tag":289,"props":981,"children":982},{"style":602},[983],{"type":58,"value":860},{"type":52,"tag":289,"props":985,"children":986},{"style":452},[987],{"type":58,"value":979},{"type":52,"tag":289,"props":989,"children":990},{"style":602},[991],{"type":58,"value":860},{"type":52,"tag":289,"props":993,"children":994},{"style":452},[995],{"type":58,"value":979},{"type":52,"tag":289,"props":997,"children":998},{"style":602},[999],{"type":58,"value":860},{"type":52,"tag":289,"props":1001,"children":1002},{"style":452},[1003],{"type":58,"value":1004}," [实验数据：seir.py 输出] ",{"type":52,"tag":289,"props":1006,"children":1007},{"style":602},[1008],{"type":58,"value":906},{"type":52,"tag":289,"props":1010,"children":1012},{"class":385,"line":1011},13,[1013],{"type":52,"tag":289,"props":1014,"children":1015},{"emptyLinePlaceholder":506},[1016],{"type":58,"value":509},{"type":52,"tag":289,"props":1018,"children":1020},{"class":385,"line":1019},14,[1021,1025],{"type":52,"tag":289,"props":1022,"children":1023},{"style":602},[1024],{"type":58,"value":780},{"type":52,"tag":289,"props":1026,"children":1027},{"style":399},[1028],{"type":58,"value":1029},"解读\n",{"type":52,"tag":289,"props":1031,"children":1033},{"class":385,"line":1032},15,[1034,1038],{"type":52,"tag":289,"props":1035,"children":1036},{"style":602},[1037],{"type":58,"value":793},{"type":52,"tag":289,"props":1039,"children":1040},{"style":452},[1041],{"type":58,"value":1042}," 在以上假设下，推演显示 \u003C峰值区间与达峰时间区间>；这些是条件推演，\n",{"type":52,"tag":289,"props":1044,"children":1046},{"class":385,"line":1045},16,[1047],{"type":52,"tag":289,"props":1048,"children":1049},{"style":452},[1050],{"type":58,"value":1051},"  不构成疫情预测。\n",{"type":52,"tag":289,"props":1053,"children":1055},{"class":385,"line":1054},17,[1056,1060],{"type":52,"tag":289,"props":1057,"children":1058},{"style":602},[1059],{"type":58,"value":793},{"type":52,"tag":289,"props":1061,"children":1062},{"style":452},[1063],{"type":58,"value":1064}," 敏感性：\u003C哪个参数最影响结果>；结构对比：\u003CSIR vs SEIR 差异>。\n",{"type":52,"tag":289,"props":1066,"children":1068},{"class":385,"line":1067},18,[1069],{"type":52,"tag":289,"props":1070,"children":1071},{"emptyLinePlaceholder":506},[1072],{"type":58,"value":509},{"type":52,"tag":289,"props":1074,"children":1076},{"class":385,"line":1075},19,[1077,1081],{"type":52,"tag":289,"props":1078,"children":1079},{"style":602},[1080],{"type":58,"value":780},{"type":52,"tag":289,"props":1082,"children":1083},{"style":399},[1084],{"type":58,"value":1085},"局限\n",{"type":52,"tag":289,"props":1087,"children":1089},{"class":385,"line":1088},20,[1090,1094],{"type":52,"tag":289,"props":1091,"children":1092},{"style":602},[1093],{"type":58,"value":793},{"type":52,"tag":289,"props":1095,"children":1096},{"style":452},[1097],{"type":58,"value":1098}," \u003C逐条列出简化假设与数据缺口>\n",{"type":52,"tag":61,"props":1100,"children":1102},{"id":1101},"本技能不做什么",[1103],{"type":58,"value":1101},{"type":52,"tag":188,"props":1105,"children":1106},{},[1107,1117,1127,1137,1154],{"type":52,"tag":97,"props":1108,"children":1109},{},[1110,1115],{"type":52,"tag":73,"props":1111,"children":1112},{},[1113],{"type":58,"value":1114},"不做参数拟合",{"type":58,"value":1116},"：不用观测数据估计参数（最小二乘\u002F极大似然\u002FMCMC\n交给计算类 skill 或专业环境）；本技能只做前向推演。",{"type":52,"tag":97,"props":1118,"children":1119},{},[1120,1125],{"type":52,"tag":73,"props":1121,"children":1122},{},[1123],{"type":58,"value":1124},"不做 SEIRS\u002FSEIRV\u002F年龄结构\u002F随机模型",{"type":58,"value":1126},"：脚本只实现 SIR 与 SEIR；\n需要更复杂结构时说明理由并转专业工具。",{"type":52,"tag":97,"props":1128,"children":1129},{},[1130,1135],{"type":52,"tag":73,"props":1131,"children":1132},{},[1133],{"type":58,"value":1134},"不做实时预测与对外发布",{"type":58,"value":1136},"：推演结果对外使用属 guardrail 第 8 条\n危险操作，必须先确认；任何\"疫情预测\"措辞在产物中禁用。",{"type":52,"tag":97,"props":1138,"children":1139},{},[1140,1145,1147,1152],{"type":52,"tag":73,"props":1141,"children":1142},{},[1143],{"type":58,"value":1144},"不编造病种参数",{"type":58,"value":1146},"：潜伏期\u002F传染期\u002FR0 的具体数值必须有来源；\n模型印象值标 ",{"type":52,"tag":138,"props":1148,"children":1150},{"className":1149},[],[1151],{"type":58,"value":143},{"type":58,"value":1153}," 且只用于演示性情景。",{"type":52,"tag":97,"props":1155,"children":1156},{},[1157,1162],{"type":52,"tag":73,"props":1158,"children":1159},{},[1160],{"type":58,"value":1161},"不做政策建议",{"type":58,"value":1163},"：干预参数（如β下降多少对应何种措施）的映射超出\n本技能证据范围，不替用户把情景翻译成决策。",{"type":52,"tag":61,"props":1165,"children":1167},{"id":1166},"收尾与下一步",[1168],{"type":58,"value":1166},{"type":52,"tag":93,"props":1170,"children":1171},{},[1172,1177,1182],{"type":52,"tag":97,"props":1173,"children":1174},{},[1175],{"type":58,"value":1176},"汇总：跑了几个情景、峰值与达峰时间的区间、最敏感的参数。",{"type":52,"tag":97,"props":1178,"children":1179},{},[1180],{"type":58,"value":1181},"指向产物（JSON\u002FCSV 序列、情景报告）并登记 provenance。",{"type":52,"tag":97,"props":1183,"children":1184},{},[1185,1187],{"type":58,"value":1186},"建议下一步：\n",{"type":52,"tag":188,"props":1188,"children":1189},{},[1190,1195,1208],{"type":52,"tag":97,"props":1191,"children":1192},{},[1193],{"type":58,"value":1194},"需要拟合真实数据或更复杂模型 → python-analysis 等计算类 skill；",{"type":52,"tag":97,"props":1196,"children":1197},{},[1198,1200,1206],{"type":58,"value":1199},"结果写入报告 → ",{"type":52,"tag":138,"props":1201,"children":1203},{"className":1202},[],[1204],{"type":58,"value":1205},"epi-writing",{"type":58,"value":1207},"（附参数三件套与情景表）；",{"type":52,"tag":97,"props":1209,"children":1210},{},[1211],{"type":58,"value":1212},"参数来源需补文献 → literature-search。",{"type":52,"tag":1214,"props":1215,"children":1216},"style",{},[1217],{"type":58,"value":1218},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1220,"total":1421},[1221,1245,1261,1280,1295,1315,1333,1351,1366,1382,1401,1411],{"slug":1222,"name":1222,"fn":1223,"description":1224,"org":1225,"tags":1226,"stars":1242,"repoUrl":1243,"updatedAt":1244},"android-native-dev","develop Android native applications","Android native application development and UI design guide. Covers Material Design 3, Kotlin\u002FCompose development, project configuration, accessibility, and build troubleshooting. Read this before Android native application development.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1227,1230,1233,1236,1239],{"name":1228,"slug":1229,"type":16},"Accessibility","accessibility",{"name":1231,"slug":1232,"type":16},"Android","android",{"name":1234,"slug":1235,"type":16},"Kotlin","kotlin",{"name":1237,"slug":1238,"type":16},"Mobile","mobile",{"name":1240,"slug":1241,"type":16},"UI Components","ui-components",13030,"https:\u002F\u002Fgithub.com\u002FMiniMax-AI\u002Fskills","2026-07-13T06:16:54.247834",{"slug":1246,"name":1246,"fn":1247,"description":1248,"org":1249,"tags":1250,"stars":1242,"repoUrl":1243,"updatedAt":1260},"buddy-sings","generate singing performances for AI companions","Use when user wants their Claude Code pet (\u002Fbuddy) to sing a song. Triggers on any request that combines the concept of their Claude Code buddy, pet, or companion with singing or music. Supports multilingual triggers — match equivalent phrases in any language.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1251,1254,1257],{"name":1252,"slug":1253,"type":16},"Agents","agents",{"name":1255,"slug":1256,"type":16},"Audio","audio",{"name":1258,"slug":1259,"type":16},"Creative","creative","2026-07-13T06:16:35.130644",{"slug":1262,"name":1262,"fn":1263,"description":1264,"org":1265,"tags":1266,"stars":1242,"repoUrl":1243,"updatedAt":1279},"color-font-skill","select color palettes and font pairings","Choose presentation-ready color palettes and font pairings for PPT\u002Fdesign tasks. Use when users ask for visual theme choices, brand-safe palettes, or font recommendations. Triggers include: 配色, 色板, 字体, color palette, font, PPT配色, 字体搭配.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1267,1270,1273,1276],{"name":1268,"slug":1269,"type":16},"Design","design",{"name":1271,"slug":1272,"type":16},"Presentations","presentations",{"name":1274,"slug":1275,"type":16},"Themes","themes",{"name":1277,"slug":1278,"type":16},"Typography","typography","2026-07-13T06:17:02.785587",{"slug":1281,"name":1281,"fn":1282,"description":1283,"org":1284,"tags":1285,"stars":1242,"repoUrl":1243,"updatedAt":1294},"design-style-skill","select visual design systems for presentations","Select a consistent visual design system for PPT slides using radius\u002Fspacing style recipes. Use when users ask for overall style direction or component styling consistency. Includes Sharp\u002FSoft\u002FRounded\u002FPill recipes, component mappings, typography\u002Fspacing rules, and mixing guidance. Triggers: 风格, style, radius, spacing, 圆角, 间距, PPT风格, 视觉风格, design style, component style.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1286,1287,1290,1293],{"name":1268,"slug":1269,"type":16},{"name":1288,"slug":1289,"type":16},"Design System","design-system",{"name":1291,"slug":1292,"type":16},"PowerPoint","powerpoint",{"name":1271,"slug":1272,"type":16},"2026-07-13T06:17:10.398389",{"slug":1296,"name":1296,"fn":1297,"description":1298,"org":1299,"tags":1300,"stars":1242,"repoUrl":1243,"updatedAt":1314},"flutter-dev","build cross-platform apps with Flutter","Flutter cross-platform development guide covering widget patterns, Riverpod\u002FBloc state management, GoRouter navigation, performance optimization, and platform-specific implementations. Includes const optimization, responsive layouts, testing strategies, and DevTools profiling.\nUse when: building Flutter apps, implementing state management (Riverpod\u002FBloc), setting up GoRouter navigation, creating custom widgets, optimizing performance, writing widget tests, cross-platform development.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1301,1304,1307,1308,1311],{"name":1302,"slug":1303,"type":16},"Dart","dart",{"name":1305,"slug":1306,"type":16},"Flutter","flutter",{"name":1237,"slug":1238,"type":16},{"name":1309,"slug":1310,"type":16},"Performance","performance",{"name":1312,"slug":1313,"type":16},"State Management","state-management","2026-07-13T06:16:36.626679",{"slug":1316,"name":1316,"fn":1317,"description":1318,"org":1319,"tags":1320,"stars":1242,"repoUrl":1243,"updatedAt":1332},"frontend-dev","build visually striking frontend web pages","Full-stack frontend development combining premium UI design, cinematic animations,\nAI-generated media assets, persuasive copywriting, and visual art. Builds complete,\nvisually striking web pages with real media, advanced motion, and compelling copy.\nUse when: building landing pages, marketing sites, product pages, dashboards,\ngenerating media assets (image\u002Fvideo\u002Faudio\u002Fmusic), writing conversion copy,\ncreating generative art, or implementing cinematic scroll animations.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1321,1324,1325,1326,1329],{"name":1322,"slug":1323,"type":16},"Animation","animation",{"name":1258,"slug":1259,"type":16},{"name":1268,"slug":1269,"type":16},{"name":1327,"slug":1328,"type":16},"Frontend","frontend",{"name":1330,"slug":1331,"type":16},"Web Development","web-development","2026-07-13T06:16:39.108827",{"slug":1334,"name":1334,"fn":1335,"description":1336,"org":1337,"tags":1338,"stars":1242,"repoUrl":1243,"updatedAt":1350},"fullstack-dev","build full-stack web applications","Full-stack backend architecture and frontend-backend integration guide.\nTRIGGER when: building a full-stack app, creating REST API with frontend, scaffolding backend service,\nbuilding todo app, building CRUD app, building real-time app, building chat app,\nExpress + React, Next.js API, Node.js backend, Python backend, Go backend,\ndesigning service layers, implementing error handling, managing config\u002Fauth,\nsetting up API clients, implementing auth flows, handling file uploads,\nadding real-time features (SSE\u002FWebSocket), hardening for production.\nDO NOT TRIGGER when: pure frontend UI work, pure CSS\u002Fstyling, database schema only.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1339,1342,1343,1346,1349],{"name":1340,"slug":1341,"type":16},"Backend","backend",{"name":1327,"slug":1328,"type":16},{"name":1344,"slug":1345,"type":16},"Full-stack","full-stack",{"name":1347,"slug":1348,"type":16},"REST API","rest-api",{"name":1330,"slug":1331,"type":16},"2026-07-13T06:16:43.219005",{"slug":1352,"name":1352,"fn":1353,"description":1354,"org":1355,"tags":1356,"stars":1242,"repoUrl":1243,"updatedAt":1365},"gif-sticker-maker","create animated GIF stickers from photos","Convert photos (people, pets, objects, logos) into 4 animated GIF stickers with captions.\nUse when: user wants to create cartoon stickers, GIF expressions, emoji packs, animated avatars,\nor convert photos to Funko Pop \u002F Pop Mart blind box style animations.\nTriggers: sticker, GIF, cartoon, emoji, expression pack, avatar animation.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1357,1358,1359,1362],{"name":1322,"slug":1323,"type":16},{"name":1258,"slug":1259,"type":16},{"name":1360,"slug":1361,"type":16},"Images","images",{"name":1363,"slug":1364,"type":16},"Media","media","2026-07-13T06:16:51.74461",{"slug":1367,"name":1367,"fn":1368,"description":1369,"org":1370,"tags":1371,"stars":1242,"repoUrl":1243,"updatedAt":1381},"ios-application-dev","develop iOS applications with SwiftUI and UIKit","iOS application development guide covering UIKit, SnapKit, and SwiftUI. Includes touch targets, safe areas, navigation patterns, Dynamic Type, Dark Mode, accessibility, collection views, common UI components, and SwiftUI design guidelines. For detailed references on specific topics, see the reference files.\nUse when: developing iOS apps, implementing UI, reviewing iOS code, working with UIKit\u002FSnapKit\u002FSwiftUI layouts, building iPhone interfaces, Swift mobile development, Apple HIG compliance, iOS accessibility implementation.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1372,1373,1376,1377,1380],{"name":1228,"slug":1229,"type":16},{"name":1374,"slug":1375,"type":16},"iOS","ios",{"name":1237,"slug":1238,"type":16},{"name":1378,"slug":1379,"type":16},"SwiftUI","swiftui",{"name":1240,"slug":1241,"type":16},"2026-07-13T06:16:55.686092",{"slug":1383,"name":1383,"fn":1384,"description":1385,"org":1386,"tags":1387,"stars":1242,"repoUrl":1243,"updatedAt":1400},"minimax-docx","create and edit DOCX documents","Professional DOCX document creation, editing, and formatting using OpenXML SDK (.NET). Three pipelines: (A) create new documents from scratch, (B) fill\u002Fedit content in existing documents, (C) apply template formatting with XSD validation gate-check. MUST use this skill whenever the user wants to produce, modify, or format a Word document — including when they say \"write a report\", \"draft a proposal\", \"make a contract\", \"fill in this form\", \"reformat to match this template\", or any task whose final output is a .docx file. Even if the user doesn't mention \"docx\" explicitly, if the task implies a printable\u002Fformal document, use this skill.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1388,1391,1394,1397],{"name":1389,"slug":1390,"type":16},"Documents","documents",{"name":1392,"slug":1393,"type":16},"DOCX","docx",{"name":1395,"slug":1396,"type":16},"Office","office",{"name":1398,"slug":1399,"type":16},"Templates","templates","2026-07-13T06:16:40.461868",{"slug":1402,"name":1402,"fn":1403,"description":1404,"org":1405,"tags":1406,"stars":1242,"repoUrl":1243,"updatedAt":1410},"minimax-music-gen","generate music and audio tracks","Use when user wants to generate music, songs, or audio tracks. Triggers on any request involving music creation, song writing, lyrics generation, audio production, or covers. Also triggers when user provides lyrics and wants them turned into a song, or describes a mood\u002Fscene and wants background music. Supports multilingual triggers — match equivalent phrases in any language. Do NOT use for music playback of existing files, music theory questions, or music recommendation without generation.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1407,1408,1409],{"name":1255,"slug":1256,"type":16},{"name":1258,"slug":1259,"type":16},{"name":1363,"slug":1364,"type":16},"2026-07-13T06:16:50.381758",{"slug":1412,"name":1412,"fn":1413,"description":1414,"org":1415,"tags":1416,"stars":1242,"repoUrl":1243,"updatedAt":1420},"minimax-music-playlist","generate personalized music playlists","Generate personalized music playlists by analyzing the user's music taste and generation feedback history. Triggers on any request involving playlist generation, music taste profiling, or personalized music recommendations. Supports multilingual triggers — match equivalent phrases in any language.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1417,1418,1419],{"name":1255,"slug":1256,"type":16},{"name":1258,"slug":1259,"type":16},{"name":1363,"slug":1364,"type":16},"2026-07-13T06:16:57.002997",84,{"items":1423,"total":1522},[1424,1440,1457,1469,1481,1496,1510],{"slug":1425,"name":1425,"fn":1426,"description":1427,"org":1428,"tags":1429,"stars":29,"repoUrl":30,"updatedAt":1439},"article2tasks","convert articles into Dida365 tasks","将技术周报（批量）或单篇文章整理为滴答清单待办事项，自动分类到已有清单并打标签，全程通过 MCP 写入，跨平台无需任何本地脚本。当用户提供周报内容、文章链接，或提及「周报」「文章汇总」「添加到滴答」时，使用此技能。",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1430,1433,1436],{"name":1431,"slug":1432,"type":16},"Content Creation","content-creation",{"name":1434,"slug":1435,"type":16},"MCP","mcp",{"name":1437,"slug":1438,"type":16},"Task Management","task-management","2026-08-18T03:49:38.751656",{"slug":1441,"name":1441,"fn":1442,"description":1443,"org":1444,"tags":1445,"stars":29,"repoUrl":30,"updatedAt":1456},"citation-verify","verify academic citation accuracy","当用户需要核验参考文献真实性、检查 BibTeX 条目是否有误、审查论文引用是否可靠、 排查 bibliography 中的错误条目、验证 DOI 与题录是否对得上、发现参考文献张冠李戴、 做投稿前引用体检或审稿引用抽查时使用。同义场景：参考文献核验、引用核查、 文献真伪鉴别、bib 文件检查、参考文献纠错、引用元数据比对、 \"帮我看看这些参考文献有没有问题\"\"这个 DOI 对得上吗\"\"有没有编造的引用\"。",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1446,1449,1452,1455],{"name":1447,"slug":1448,"type":16},"Code Analysis","code-analysis",{"name":1450,"slug":1451,"type":16},"Documentation","documentation",{"name":1453,"slug":1454,"type":16},"QA","qa",{"name":14,"slug":15,"type":16},"2026-08-21T03:59:17.615117",{"slug":1458,"name":1458,"fn":1459,"description":1460,"org":1461,"tags":1462,"stars":29,"repoUrl":30,"updatedAt":1468},"claim-check","verify research claims against evidence","当需要检查论文草稿中的事实性论断是否有证据支撑、做引用-论断对齐审计、 找出没有出处的陈述、核对文内引用是否真能支撑所在句子时加载使用； 通常由 evidence-loop 或审稿类流程调用，不直接面向用户。同义场景： 论断支撑检查、claim 支撑审计、无支撑断言排查、证据覆盖检查、 引用与论断对齐、\"这段话有没有文献撑\"\"哪些说法没有引用\"。",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1463,1466,1467],{"name":1464,"slug":1465,"type":16},"Audit","audit",{"name":1447,"slug":1448,"type":16},{"name":14,"slug":15,"type":16},"2026-08-21T03:59:02.154691",{"slug":1470,"name":1470,"fn":1471,"description":1472,"org":1473,"tags":1474,"stars":29,"repoUrl":30,"updatedAt":1480},"cn-literature","search and organize Chinese academic literature","当用户需要检索或整理中文文献（CNKI 知网、万方、维普、超星）、把中文数据库 导出的题录文件解析成统一 PaperDocument、或把中文文献与英文检索结果合并去重 时使用。同义场景：中文文献检索、知网导出题录解析、万方检索、Refworks\u002FEndNote 格式转换、\"帮我把这份 CNKI 导出的 txt 整理成文献表\"\"中文核心期刊上关于 X 有哪些研究\"。",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1475,1478,1479],{"name":1476,"slug":1477,"type":16},"Data Cleaning","data-cleaning",{"name":1389,"slug":1390,"type":16},{"name":14,"slug":15,"type":16},"2026-08-21T03:59:12.24404",{"slug":1482,"name":1482,"fn":1483,"description":1484,"org":1485,"tags":1486,"stars":29,"repoUrl":30,"updatedAt":1495},"cold-start-interview","initialize research profiles via interviews","当用户首次使用科研工作台、研究画像仍为空白，或用户主动说\"先做个访谈\"\"重新设置我的研究信息\"\"初始化画像\"\"cold start\"\"重新 interview\"时使用。本技能通过结构化访谈收集用户的研究领域、常用数据源、算力环境、写作语言与目标期刊、引用格式、团队协作规范、伦理合规要求，并把回答以散文形式写回插件根目录 CLAUDE.md 的研究画像部分。支持 quick（2 分钟 5 问）、full（完整 15 问）、--redo（推翻重谈）、--check（只检查画像完整度不提问）四种模式，支持中断后续谈。同义触发场景：新用户引导、首次配置、研究背景调查、填研究档案、更新个人研究信息、check 画像。",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1487,1488,1491,1494],{"name":1450,"slug":1451,"type":16},{"name":1489,"slug":1490,"type":16},"Onboarding","onboarding",{"name":1492,"slug":1493,"type":16},"Productivity","productivity",{"name":14,"slug":15,"type":16},"2026-08-21T03:59:58.579998",{"slug":1497,"name":1497,"fn":1498,"description":1499,"org":1500,"tags":1501,"stars":29,"repoUrl":30,"updatedAt":1509},"customize","customize research guardrails and settings","当用户想修改研究画像或共享守则（guardrails）时使用：换了研究领域、换了目标期刊、换了引用格式、换了算力环境、合规要求变化、想调整某条 guardrail 的严格程度。同义触发场景：改一下画像、更新我的研究信息、修改 CLAUDE.md、调整设置、换引用格式、customize、个性化配置、改规则、重新配置工作台。本技能是画像与守则的唯一维护入口：读取 CLAUDE.md、与用户逐项确认修改点、写回文件、并明确告知哪些 skill 的行为会因此变化。",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1502,1505,1508],{"name":1503,"slug":1504,"type":16},"Best Practices","best-practices",{"name":1506,"slug":1507,"type":16},"Configuration","configuration",{"name":14,"slug":15,"type":16},"2026-08-21T03:59:35.867596",{"slug":1511,"name":1511,"fn":1512,"description":1513,"org":1514,"tags":1515,"stars":29,"repoUrl":30,"updatedAt":1521},"dida365","manage Dida365 tasks and habits","Manage 滴答清单 (Dida365 China) tasks, lists, habits, focus records and countdowns through the official Dida365 MCP server at mcp.dida365.com. Use when the user asks to check, plan, create, update, complete or review tasks and habits on dida365.com, or asks about focus history and countdowns.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1516,1517,1520],{"name":1434,"slug":1435,"type":16},{"name":1518,"slug":1519,"type":16},"Scheduling","scheduling",{"name":1437,"slug":1438,"type":16},"2026-08-18T03:49:37.126385",36]