Medical Research Agent Skills is a library of agent instructions for medical and biomedical research, covering evidence analysis, study protocol design, data analysis, and academic writing. Researchers use it to guide compatible coding agents through common scientific workflows. The catalogue contains many of the library's skills and commands.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add aipoch/medical-research-skills --skill real-world-evidence-study-designergit clone --depth 1 https://github.com/aipoch/medical-research-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aipoch/medical-research-skills/real-world-evidence-study-designer)<a href="https://agentmods.dev/skills/aipoch/medical-research-skills/real-world-evidence-study-designer"><img src="https://agentmods.dev/badge/skills/aipoch/medical-research-skills/real-world-evidence-study-designer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aipoch/medical-research-skills/real-world-evidence-study-designer"><img src="https://agentmods.dev/badge/skills/aipoch/medical-research-skills/real-world-evidence-study-designer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 198 Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00102 | $0.02904 |
| Opus 5 | $0.00051 | $0.01452 |
| Sonnet 5 | $0.00020 | $0.00581 |
| Haiku 4.5 | $0.00010 | $0.00290 |
Grade A, and why
real-world-evidence-study-designer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 241 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Real-World Evidence Study Designer
You are an expert clinical epidemiology and real-world evidence study-design strategist.
Task: Convert a clinical or translational research question into a real-world evidence study blueprint that is explicit about data source fit, cohort construction logic, time zero, exposure definition, outcome windowing, censoring rules, confounder control, and target-trial-emulation discipline.
This skill is for users who need study type design / protocol framing, not a full protocol, not a manuscript, and not an unqualified causal claim. The output should show how the study would actually be structured using EHR, claims, or registry data, where the key design vulnerabilities are, and which assumptions remain unverified.
This skill must always distinguish between:
- the target clinical question versus the estimand that the available real-world data can actually support
- eligibility logic versus analytic subgroup logic
- baseline information available before or at time zero versus post-baseline information that must not be treated as baseline confounders
- incident exposure (new-user) designs versus prevalent-user designs
- treatment initiation, exposure episode construction, and treatment switching / discontinuation
- outcome ascertainment windows versus follow-up completeness assumptions
- database convenience versus design validity
- association-oriented RWE versus causal target-trial-emulation framing
- variables truly captured in the selected data source versus variables the ideal study would want but may not have
This skill must not confuse RWE study design with simple retrospective chart review, cross-sectional database description, randomized trial design, or unsupported causal inference.
Reference Module Integration
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- eval_report_real-world-evidence-study-designer_result.json 11 KB
- references/analysis-line-framework.md 649 B
- references/confounding-and-bias-control-rules.md 660 B
- references/data-source-and-capture-framework.md 948 B
- references/literature-integrity-rules.md 394 B
- references/output-section-guidance.md 366 B
- references/rwe-question-fit-rules.md 1.0 KB
- references/target-trial-emulation-rules.md 812 B
- references/time-zero-exposure-followup-rules.md 737 B
- references/workflow-step-template.md 516 B
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 241 lines · 102 tokens per session scan A 34371dbd9f2c
real-world-evidence-study-designer is a skill published in the GitHub repository aipoch/medical-research-skills (1,869 stars, last pushed today), licensed MIT. It adds 102 tokens to every session and 2,904 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
statistical-modeling
Statistical modeling and machine learning for biomarker discovery, survival analysis, classification, regression, and model interpretation.
bulk-transcriptomics
Bulk RNA-seq and microarray differential expression analysis including method selection, batch correction, and complex experimental designs.
chromatin-regulation
Chromatin regulation analysis from called peaks and count matrices — differential binding, signal summarisation, peak annotation, and scATAC-seq.
spatial-omics
Spatial transcriptomics and spatial proteomics analysis covering technology-specific workflows, spatial statistics, deconvolution, and niche analysis.
atac-seq-bam-read-alignment-processing
Use when when you have aligned ATAC-seq BAM files and need to quantify Tn5 transposase insertion patterns around specific genomic coordinates (motif sites, peaks, regulatory regions) to detect transcription factor occupancy footprints or compare chromatin accessibility between bound and unbound.
bedgraph-file-format-manipulation
Use when you have aligned ChIP-Seq reads (in BED or BEDPE format) and need to convert them into quantitative genome-wide signal tracks (coverage, p-value, or q-value scores) for downstream statistical comparison or peak detection.