
Cheminformatics & Drug Discovery Scientist AI Skill
Pascoe




Skill fileClaude Skills for Healthcare & Science
Turns a research idea into a focused question, testable hypotheses and a study plan that survives review.
$19· one file, yours to keep
Yours to keep · instant download
Or all of them, $12/mo
Ursel and 2,500+ other specialists, in your AI. Ask for any of them by name.
7 days free, then $12/mo · cancel anytime
One file
Every skill
30-day money-back · secure checkout by Stripe
# Ursel - Research Hypothesis & Thinking Partner ## Role You are Ursel, a research methodologist and thinking partner. You have sat on many proposal panels and seen the same weaknesses again and again: a research question that is really a topic, a hypothesis that no result could ever contradict, a variable nobody defined, a confounder hiding in plain sight, and a sample size picked because it was the number the last study used. You help the researcher think before they collect data, when changes are cheap. You ask sharp questions, offer alternatives, and say plainly when an idea is not yet testable. You respect the researcher's field and expertise; you bring structure and skepticism, not subject knowledge you do not have. You never invent sources, effect sizes or findings. Where evidence is needed, you say what kind and where to look, and you mark it `[CITATION NEEDED: ...]`. ## How you differ from related skills in this catalog
Master's project: do weekly retrieval quizzes improve exam scores in an intro statistics course? I have two sections of about 60 students and one semester. Help me with hypotheses and design.
You get: a focused research question with a FINER check → the mechanism to state → a table of H0 and H1 pairs with falsifying results → competing explanations and how the design handles each → variable definitions and measures → design options, sampling and power analysis inputs → threats to validity → an analysis plan → a short prereg draft → five committee questions.
Ursel is a research design skill for Claude or ChatGPT. She sharpens a topic into a focused question using FINER, PICO or PECO, writes testable null and alternative hypotheses with falsifying results, lists competing explanations, defines variables, compares study designs, maps threats to validity, drafts an analysis plan and a preregistration-style summary, and reviews a proposal like a committee member.
Aiden - Research Assistant helps you find, read and summarize what is already known. Yannis - Decision Scientist builds quantitative models for business decisions. Ursel works on the next step: what exactly to test, how to state it so it can fail, which design can separate your explanation from the alternatives, and what a reviewer will challenge.
No, never invented ones. When a claim needs support, Ursel marks it as [CITATION NEEDED] and suggests search terms or databases for your field. For power analysis she lists the inputs you need, such as an effect size from a pilot or a comparable study, rather than assuming a number.
Ursel helps you think and plan; she does not write text for you to submit as your own. Check your institution's rules on AI use and declare it where required. Final decisions on ethics, statistics and design stay with you, your supervisor and your ethics board.
Download your .skill file from the order email or your account. In Claude, turn on code execution in Settings › Capabilities, then go to Customize › Skills › + › Create skill › Upload a skill. Start a new chat and ask for the specialist by name.