[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EMJS30M8V8ZB0Y38RB4R5G":3,"skill-same:01M7EMJS30M8V8ZB0Y38RB4R5G":47},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M7EMJS30M8V8ZB0Y38RB4R5G","clinical-trial-design",null,"Designing clinical trials — phases, randomization, endpoints, sample size, and CONSORT-compliant reporting.","TanNhatCMS\u002Fjarvis","TanNhatCMS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113796420?v=4","01M753MV17KG5CD3152G3WQEQY","KietNT","01M7EKEDYVYQ6PWVJJMZ911VMX",0,"MIT","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-clinical-trial-design\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: clinical-trial-design\ndescription: Designing clinical trials — phases, randomization, endpoints, sample size, and CONSORT-compliant reporting.\ncategory: scientific\n---\n\n## Overview\n\nclinical-trial-design covers the architecture of interventional clinical studies: choosing the\nphase and design (parallel, crossover, factorial, adaptive), defining endpoints, randomizing and\nblinding, sizing the sample, and planning analyses that regulators and journals accept. A trial's\ncredibility is fixed at the design stage — no analysis rescues a trial that measured the wrong\nthing in the wrong population.\n\n## When to use\n\n- Choosing trial phase and design for a new intervention.\n- Defining primary\u002Fsecondary endpoints and estimands.\n- Randomization schemes: simple, blocked, stratified, minimization.\n- Blinding: who, how, and when it's impossible (surgical\u002Fbehavioral trials).\n- Sample-size calculation and interim analyses with alpha spending.\n- Adaptive designs: group sequential, sample-size re-estimation, platform trials.\n- Writing protocols and CONSORT-compliant reports.\n\n## Core concepts\n\n- **Phases.** Phase I: safety\u002Ftolerability (often healthy volunteers, dose escalation). Phase II:\n  preliminary efficacy + dose finding. Phase III: confirmatory efficacy vs standard of care\n  (large, randomized, the basis for approval). Phase IV: post-marketing surveillance. Skipping\n  phases is occasionally justified, never casual.\n- **Estimands (ICH E9 R1).** Define precisely what treatment effect you estimate: population,\n  treatment conditions, endpoint, handling of intercurrent events (discontinuation, rescue\n  medication), summary measure. \"Intention-to-treat\" is an estimand choice, not a default virtue —\n  state it.\n- **Endpoints.** Primary: one (or a co-primary pair), clinically meaningful, measured reliably.\n  Surrogate endpoints (biomarkers) are acceptable only when validated — lowering a lab value is\n  not the same as helping patients. Patient-reported outcomes belong in most modern trials.\n- **Randomization.** Blocked randomization keeps arms balanced over time; stratification balances\n  key prognostic factors (site, disease severity); minimization for many strata. Use a central,\n  concealed system — envelope randomization is a known failure point.\n- **Blinding.** Double-blind where possible; when impossible (surgery, psychotherapy), blind the\n  outcome assessors and statisticians at minimum, and use objective endpoints. Assess blinding\n  success — failed blinding with subjective endpoints is a major bias source.\n- **Sample size.** Powered on the primary endpoint's minimum clinically important difference,\n  with realistic variance and dropout inflation (10-20%). Underpowered trials are unethical —\n  they expose patients to risk for uninterpretable results.\n- **Interim analyses.** Group-sequential designs with alpha-spending functions (O'Brien-Fleming,\n  Pocock) allow early stopping for efficacy or futility. Unplanned peeking at the data inflates\n  Type I error — all interim looks must be prespecified with a data monitoring committee.\n- **Design variants.** Crossover (each patient gets both treatments — efficient, but only for\n  stable chronic conditions without carryover); factorial (two interventions at once, tests\n  interaction); non-inferiority (margin must be clinically justified, not statistically\n  convenient); platform\u002Fadaptive trials (shared infrastructure, dropping arms — efficient but\n  operationally demanding).\n\n## Practical workflow\n\n1. **PICO.** Population, Intervention, Comparator, Outcome — one sentence that the whole trial\n   hangs on. If the comparator isn't the real clinical alternative, the trial answers the wrong\n   question.\n2. **Estimand and endpoints.** Primary endpoint, intercurrent-event strategy, analysis\n   population. Register on ClinicalTrials.gov before enrollment.\n3. **Design choice.** Parallel-group default; crossover\u002Ffactorial\u002Fadaptive only with justification.\n4. **Size it.** Power 80-90% on the MCID; inflate for dropout; document assumptions.\n5. **Randomize and blind.** Central system, stratified blocks; blinding plan with fallback for\n   unblindable interventions.\n6. **Monitor.** Independent DMC, prespecified interim analyses, safety stopping rules.\n7. **Analyze as planned.** ITT primary; per-protocol as sensitivity; handle missing data with\n   prespecified methods (not last-observation-carried-forward).\n8. **Report CONSORT.** Flow diagram, baseline table, primary result with effect size + CI,\n   harms by arm. Publish regardless of outcome — publication bias kills evidence synthesis.\n\n## Common pitfalls\n\n- Surrogate primary endpoints without validation (the drug works on the biomarker, not the patient).\n- Underpowered trials (\"exploratory\" as an excuse for n=30 confirmatory claims).\n- Unplanned interim looks inflating false positives.\n- Per-protocol presented as primary (breaks randomization).\n- Non-inferiority margins chosen to guarantee success.\n- Comparator that's a straw man (placebo where an active standard exists).\n- Selective outcome reporting — publishing a different primary endpoint than registered.\n","## Overview\n\nclinical-trial-design covers the architecture of interventional clinical studies: choosing the\nphase and design (parallel, crossover, factorial, adaptive), defining endpoints, randomizing and\nblinding, sizing the sample, and planning analyses that regulators and journals accept. A trial's\ncredibility is fixed at the design stage — no analysis rescues a trial that measured the wrong\nthing in the wrong population.\n\n## When to use\n\n- Choosing trial phase and design for a new intervention.\n- Defining primary\u002Fsecondary endpoints and estimands.\n- Randomization schemes: simple, blocked, stratified, minimization.\n- Blinding: who, how, and when it's impossible (surgical\u002Fbehavioral trials).\n- Sample-size calculation and interim analyses with alpha spending.\n- Adaptive designs: group sequential, sample-size re-estimation, platform trials.\n- Writing protocols and CONSORT-compliant reports.\n\n## Core concepts\n\n- **Phases.** Phase I: safety\u002Ftolerability (often healthy volunteers, dose escalation). Phase II:\n  preliminary efficacy + dose finding. Phase III: confirmatory efficacy vs standard of care\n  (large, randomized, the basis for approval). Phase IV: post-marketing surveillance. Skipping\n  phases is occasionally justified, never casual.\n- **Estimands (ICH E9 R1).** Define precisely what treatment effect you estimate: population,\n  treatment conditions, endpoint, handling of intercurrent events (discontinuation, rescue\n  medication), summary measure. \"Intention-to-treat\" is an estimand choice, not a default virtue —\n  state it.\n- **Endpoints.** Primary: one (or a co-primary pair), clinically meaningful, measured reliably.\n  Surrogate endpoints (biomarkers) are acceptable only when validated — lowering a lab value is\n  not the same as helping patients. Patient-reported outcomes belong in most modern trials.\n- **Randomization.** Blocked randomization keeps arms balanced over time; stratification balances\n  key prognostic factors (site, disease severity); minimization for many strata. Use a central,\n  concealed system — envelope randomization is a known failure point.\n- **Blinding.** Double-blind where possible; when impossible (surgery, psychotherapy), blind the\n  outcome assessors and statisticians at minimum, and use objective endpoints. Assess blinding\n  success — failed blinding with subjective endpoints is a major bias source.\n- **Sample size.** Powered on the primary endpoint's minimum clinically important difference,\n  with realistic variance and dropout inflation (10-20%). Underpowered trials are unethical —\n  they expose patients to risk for uninterpretable results.\n- **Interim analyses.** Group-sequential designs with alpha-spending functions (O'Brien-Fleming,\n  Pocock) allow early stopping for efficacy or futility. Unplanned peeking at the data inflates\n  Type I error — all interim looks must be prespecified with a data monitoring committee.\n- **Design variants.** Crossover (each patient gets both treatments — efficient, but only for\n  stable chronic conditions without carryover); factorial (two interventions at once, tests\n  interaction); non-inferiority (margin must be clinically justified, not statistically\n  convenient); platform\u002Fadaptive trials (shared infrastructure, dropping arms — efficient but\n  operationally demanding).\n\n## Practical workflow\n\n1. **PICO.** Population, Intervention, Comparator, Outcome — one sentence that the whole trial\n   hangs on. If the comparator isn't the real clinical alternative, the trial answers the wrong\n   question.\n2. **Estimand and endpoints.** Primary endpoint, intercurrent-event strategy, analysis\n   population. Register on ClinicalTrials.gov before enrollment.\n3. **Design choice.** Parallel-group default; crossover\u002Ffactorial\u002Fadaptive only with justification.\n4. **Size it.** Power 80-90% on the MCID; inflate for dropout; document assumptions.\n5. **Randomize and blind.** Central system, stratified blocks; blinding plan with fallback for\n   unblindable interventions.\n6. **Monitor.** Independent DMC, prespecified interim analyses, safety stopping rules.\n7. **Analyze as planned.** ITT primary; per-protocol as sensitivity; handle missing data with\n   prespecified methods (not last-observation-carried-forward).\n8. **Report CONSORT.** Flow diagram, baseline table, primary result with effect size + CI,\n   harms by arm. Publish regardless of outcome — publication bias kills evidence synthesis.\n\n## Common pitfalls\n\n- Surrogate primary endpoints without validation (the drug works on the biomarker, not the patient).\n- Underpowered trials (\"exploratory\" as an excuse for n=30 confirmatory claims).\n- Unplanned interim looks inflating false positives.\n- Per-protocol presented as primary (breaks randomization).\n- Non-inferiority margins chosen to guarantee success.\n- Comparator that's a straw man (placebo where an active standard exists).\n- Selective outcome reporting — publishing a different primary endpoint than registered.\n",{"name":5,"category":24,"description":7},"scientific","main","af388a77c2fe5e0b4fd879cca4300a0ffa9a6579","skills\u002Fams-clinical-trial-design",1,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis","https:\u002F\u002Fgithub.com\u002FTanNhatCMS","https:\u002F\u002Fmrkiet.name.vn","2026-11-07 14:23:08",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add TanNhatCMS\u002Fjarvis --skill clinical-trial-design",{"tool":39,"command":40},"bunx","bunx skills add TanNhatCMS\u002Fjarvis --skill clinical-trial-design",{"tool":42,"command":43},"pnpm","pnpm dlx skills add TanNhatCMS\u002Fjarvis --skill clinical-trial-design","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-clinical-trial-design\u002FSKILL.md",true,[],[48,55,63,70,77,84],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7EWYJR0J8DARY7E59Z6Y9MV","web-scrape","Turn web pages into LLM-ready markdown via the Firecrawl API (renders JS, anti-bot handling). Use when browser_open\u002Ffetch fails on JS-heavy pages, or when you need clean markdown\u002Fstructured extraction from a site. Requires FIRECRAWL_API_KEY.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fweb-scrape\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":59,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7EWYBN5P6F2AJG3FR24T9BP","design-taste-frontend","tasteskill v2, the anti-slop frontend skill. Art direction for websites, distilled from a study of 247 curated gallery sites (One Page Love, Awwwards Sites of the Day). Deals researched directions (idea, composition, type, palette, photographs, section blueprints) from a library, enforces a strict no-slop house style with a real-browser inspector, and runs a look-and-fix loop on screenshots. Use this whenever you build, design, redesign, restyle, polish or review a landing page, marketing site, portfolio, product page, one-pager, hero or any public-facing web page, in any stack (HTML\u002FCSS, React, Next.js, Vue, Svelte, Astro, Tailwind) — also when the user says a page looks generic, templated, \"AI-made\", sloppy, or asks to make it look designed, premium, award-level or less like every other site. Not for dashboards or app screens.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7EWY8FQFNQ2GB7NZZWGSSDX","test-driven-development","Use when implementing any feature or bugfix, before writing implementation code","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-tdd\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7EWY5SHBFRH1HDHVWM9K7TV","systematic-debugging","Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-debugging\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7EWY3NM0Z0JAKS7HDZ24K66","requesting-code-review","Use when completing tasks, implementing major features, or before merging to verify work meets requirements","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-code-review\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7EWY0WRMWX4M9ZFY0WKC593","brainstorming","You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-brainstorming\u002FSKILL.md",[],[]]