Clinical Trial Protocol Skill
A clinical trial protocol stands or falls on a few linked decisions: a clear objective, a design that can answer
it, endpoints that measure it, eligibility that defines who's studied, and a statistical plan that can detect the
effect. This skill drafts a protocol synopsis that makes those decisions explicit and internally consistent, in
the structure IRBs/ethics committees and regulators expect. (For a general academic or UX study, use
research-protocol.)
Safety & compliance note: this is a drafting aid for expert review, not regulatory, medical, or
statistical sign-off. Real trials require qualified investigators, a statistician, and IRB/ethics and
regulatory approval (e.g. GCP, ICH, local law). Do not invent efficacy/safety data; mark assumptions for the
study team to set.
Working from a brief
Given "a phase II trial of drug X for condition Y", produce the full synopsis anyway — infer a defensible
design, endpoints, and eligibility appropriate to the phase and condition, and clearly label every inferred
choice as a draft assumption for the study team and statistician to confirm. Never fabricate prior data or
effect sizes; state them as placeholders to be set.
Required Inputs
Ask for these only if they aren't already provided (else infer and label as draft):
- Intervention & condition — what's being studied, in whom, and the phase.
- Objective / question — the primary question the trial must answer.
- Comparator — placebo, standard of care, or active control; and blinding.
- Outcome of interest — how benefit (and harm) will be measured.
- Constraints — known population, setting, and any regulatory context.
Output Format
Clinical Trial Protocol Synopsis: [title]
- 1. Background & rationale — the problem, prior evidence (mark placeholders), and why this trial.
- 2. Objectives — primary and secondary, each as a precise, testable statement.
- 3. Design — phase, type (RCT, etc.), allocation/randomisation, blinding, arms, and duration.
- 4. Population — setting, and explicit inclusion and exclusion criteria.
- 5. Interventions — the intervention and comparator: dose/regimen, administration, and concomitant rules.
- 6. Endpoints — primary endpoint (one, tied to the primary objective), secondary endpoints, and how/when each is measured.
- 7. Statistical considerations — analysis populations, the primary analysis, and a sample-size basis (with assumptions flagged for the statistician).
- 8. Safety — adverse-event definitions, monitoring/reporting, stopping rules, and any DSMB.
- 9. Ethics & conduct — informed consent, IRB/ethics approval, data integrity, and GCP adherence.
Close with assumptions to confirm and a reminder that a qualified investigator and statistician must own the final protocol.
Quality Checks
Anti-Patterns
Based On
Clinical research practice — objective-endpoint-analysis alignment, explicit eligibility, sample-size justification, and ICH-GCP safety/ethics structure.
Example Trigger Phrases
- "Write a clinical trial protocol."
- "Draft a study protocol synopsis."
- "Design the trial."
- "Structure endpoints, eligibility and statistics for an interventional study."
1---2name: clinical-trial-protocol3description: "Draft a clinical trial protocol synopsis with the elements regulators and IRBs expect. Use when asked to write a clinical trial protocol, a study protocol synopsis, a trial design, or to structure endpoints/eligibility/statistics for an interventional study. Produces a structured protocol synopsis — objectives, design, population with eligibility, interventions, endpoints, statistics, and safety/ethics — for expert review. (For non-clinical/UX research, use research-protocol.)"4disable-model-invocation: true # skill-optimizer5metadata:6 source: https://skillmd.com/skills/slipcast-dev/clinical-trial-protocol7---89# Clinical Trial Protocol Skill1011A clinical trial protocol stands or falls on a few linked decisions: a clear objective, a design that can answer12it, endpoints that measure it, eligibility that defines who's studied, and a statistical plan that can detect the13effect. This skill drafts a protocol synopsis that makes those decisions explicit and internally consistent, in14the structure IRBs/ethics committees and regulators expect. (For a general academic or UX study, use15[`research-protocol`](../research-protocol/SKILL.md).)1617> **Safety & compliance note:** this is a drafting aid for **expert review**, not regulatory, medical, or18> statistical sign-off. Real trials require qualified investigators, a statistician, and IRB/ethics and19> regulatory approval (e.g. GCP, ICH, local law). Do not invent efficacy/safety data; mark assumptions for the20> study team to set.2122## Working from a brief2324Given "a phase II trial of drug X for condition Y", **produce the full synopsis anyway** — infer a defensible25design, endpoints, and eligibility appropriate to the phase and condition, and clearly label every inferred26choice as a draft assumption for the study team and statistician to confirm. Never fabricate prior data or27effect sizes; state them as placeholders to be set.2829## Required Inputs3031Ask for these only if they aren't already provided (else infer and label as draft):3233- **Intervention & condition** — what's being studied, in whom, and the phase.34- **Objective / question** — the primary question the trial must answer.35- **Comparator** — placebo, standard of care, or active control; and blinding.36- **Outcome of interest** — how benefit (and harm) will be measured.37- **Constraints** — known population, setting, and any regulatory context.3839## Output Format4041### Clinical Trial Protocol Synopsis: [title]4243- **1. Background & rationale** — the problem, prior evidence (mark placeholders), and why this trial.44- **2. Objectives** — primary and secondary, each as a precise, testable statement.45- **3. Design** — phase, type (RCT, etc.), allocation/randomisation, blinding, arms, and duration.46- **4. Population** — setting, and explicit **inclusion** and **exclusion** criteria.47- **5. Interventions** — the intervention and comparator: dose/regimen, administration, and concomitant rules.48- **6. Endpoints** — **primary endpoint** (one, tied to the primary objective), secondary endpoints, and how/when each is measured.49- **7. Statistical considerations** — analysis populations, the primary analysis, and a sample-size basis (with assumptions flagged for the statistician).50- **8. Safety** — adverse-event definitions, monitoring/reporting, stopping rules, and any DSMB.51- **9. Ethics & conduct** — informed consent, IRB/ethics approval, data integrity, and GCP adherence.5253Close with **assumptions to confirm** and a reminder that a qualified investigator and statistician must own the final protocol.5455## Quality Checks5657- [ ] The primary objective, primary endpoint, and primary analysis are aligned and consistent58- [ ] There is exactly one primary endpoint, clearly measurable and time-anchored59- [ ] Inclusion/exclusion criteria are explicit and operationally checkable60- [ ] Sample-size basis is stated with its assumptions flagged for the statistician61- [ ] Safety monitoring, reporting, and stopping rules are present62- [ ] Ethics/consent/IRB and GCP elements are included; no efficacy/safety data is invented6364## Anti-Patterns6566- [ ] Do not invent prior efficacy/safety data or effect sizes — mark them as placeholders to be set67- [ ] Do not list multiple "primary" endpoints — pick one and demote the rest to secondary68- [ ] Do not let objective, endpoint, and analysis drift apart — they must answer the same question69- [ ] Do not present this as regulatory/statistical sign-off — it's a draft for expert review70- [ ] Do not omit safety monitoring and stopping rules — a protocol without them isn't approvable7172## Based On7374Clinical research practice — objective-endpoint-analysis alignment, explicit eligibility, sample-size justification, and ICH-GCP safety/ethics structure.7576## Example Trigger Phrases7778- "Write a clinical trial protocol."79- "Draft a study protocol synopsis."80- "Design the trial."81- "Structure endpoints, eligibility and statistics for an interventional study."