Skip to main content
skill CX Codex

tooluniverse-dataset-discovery — Claude Code skill: how to install

tooluniverse-dataset-discovery is a Claude Code skill that finds research datasets and checks study designs and variables for scientific questions.

Apache-2.0 254 installs ★ 1,798 updated 2026-10-24

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-dataset-discovery -a claude-code
Open on GitHub

What tooluniverse-dataset-discovery does

This skill matches research hypotheses to explicit dataset specifications before searching. First, the agent establishes the mandatory study design. If a question tracks changes over time, cross-sectional surveys are rejected immediately. They cannot establish temporal sequence. The agent then lists exposures, outcomes, and essential confounders.

Discovery spans multiple search layers. It checks re3data, domain portals like NHANES or GEO, and literature methods in PubMed. The skill filters candidate datasets by statistical power, missingness, and variable formats.

Repositories like dbGaP and EGA return study metadata only. Raw individual-level genetic data remains inaccessible without an approved data access application. Once an open dataset is selected, the agent writes Python code and loads the files directly via Bash.

What it helps with

  • map research questions to longitudinal or experimental study designs
  • find relevant public datasets across GEO, SRA, and NHANES
  • evaluate candidate studies for statistical power and necessary confounders
  • inspect metadata for restricted repositories like dbGaP and EGA

Example prompts for the agent

  • Find longitudinal datasets examining whether sleep duration predicts cognitive decline.
  • Search NHANES for dietary sugar intake variables and demographic confounders.
  • Look up public genomic datasets in GEO for gene expression profiles in asthma.

How to install tooluniverse-dataset-discovery

1Skills CLI (Vercel)

npx skills add mims-harvard/ToolUniverse --skill tooluniverse-dataset-discovery -a claude-code

2Manually: put the files in place

~/.claude/skills/tooluniverse-dataset-discovery/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/tooluniverse-dataset-discovery/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about tooluniverse-dataset-discovery

How to install tooluniverse-dataset-discovery in Claude Code?

Run npx skills add mims-harvard/ToolUniverse --skill tooluniverse-dataset-discovery -a claude-code in your terminal.

What does tooluniverse-dataset-discovery do?

It translates research questions into concrete study designs and searches cross-disciplinary repositories such as re3data, GEO, NHANES, and PubMed for matching datasets.

Can the skill directly download controlled-access genomic data?

No. Tools for repositories like dbGaP and EGA retrieve only study metadata, as raw individual genotype files require a formal institutional access request.

Source text

SKILL.md · Apache-2.0
---
name: tooluniverse-dataset-discovery
description: Find and evaluate research datasets for any scientific question. Maps research questions to required study designs (longitudinal vs cross-sectional, observational vs experimental, single-cohort vs multi-cohort). Use when the user asks 'find data about X', 'where can I get data on Y', or needs a specific cohort/survey/repository. Covers GEO, ArrayExpress, dbGaP, NHANES, UK Biobank, ClinicalTrials.gov, GWAS Catalog, and 30+ scientific repositories.
disable-model-invocation: true
---

# Dataset Discovery

## When to Use
- User asks "find me data about X" or "where can I get data on Y"
- User wants to analyze a relationship between variables
- User needs specific study designs (longitudinal, cross-sectional, experimental)
- User asks about specific surveys or cohorts

## Step 1: Understand What the Research Question Requires

Before searching, determine the **minimum data requirements**:

**Study design needed:**
- "Does X predict CHANGES in Y over time?" → longitudinal (same people measured repeatedly). Cross-sectional data CANNOT answer this — don't settle for it.
- "Is X associated with Y?" → cross-sectional is sufficient (one-time measurement)
- "Does intervention X cause outcome Y?" → experimental (clinical trial with controls)
- "What genes/proteins are involved in X?" → omics (sequencing, expression, proteomics)

**Variables needed:**
- List the specific exposure, outcome, and confounder variables
- For each variable, note the measurement type (continuous, categorical, biomarker vs self-report)
- Identify minimum confounders needed (age, sex are almost always required; domain-specific confounders depend on the question)

**Population needed:**
- Age range, geography, clinical status, sample size requirements
- Power analysis: to detect a small effect (r=0.1), you need ~800 subjects at 80% power

## Step 2: Search Strategy

Search from broadest to most specific. Use `find_tools` to discover available dataset search tools — don't rely on memorized tool names.

**Layer 1 — Cross-repository search (cast wide net):**
Search tools that index datasets across thousands of repositories. These find datasets you didn't know existed.
- Search by: research topic keywords, variable names, population descriptors
- Look for: DOI-registered datasets, repository listings, government data portals
- `re3data_search_repositories`/`re3data_get_repository` — a registry of 3,000+ research data REPOSITORIES themselves (not individual datasets), covering every discipline; use this one level up from a dataset search when the question is "what repository should I even be looking in for this kind of data" (verified live: searching returns real health-research repositories, e.g. the "EPIC study" cohort and "Pacific Islands Families Study"). Once you've identified the right repository from re3data, go search *inside* it with the domain-specific tools in Layer 2 below.

**Layer 2 — Domain-specific repositories:**
Search repositories specialized for your data type.
- Health surveys: CDC, NHANES (US; search by variable name, not topic keywords); DHS Program (`DHSProgram_search_indicators`/`_get_data`) for the international/LMIC equivalent -- fertility, maternal/child mortality, nutrition, immunization, HIV indicators across national surveys, filterable by country and year
- Genomics (open): SRA, ENA, ArrayExpress, GEO, and DDBJ (`DDBJ_search_entries`/`_get_entry`/`_get_cross_references`) -- the third INSDC archive alongside SRA/ENA, plus GEA/MetaboBank entries; also GSA (`GSA_get_accession`) for China's national genomics archive
- Genomics (controlled-access, human studies): dbGaP (`DbGaP_search_studies`/`_get_study`) and EGA (`EGA_get_study`/`_get_dataset`/`_get_study_datasets`) -- both require a separate data-access request process; these tools return study/dataset metadata, not the underlying genotype data itself
- Proteomics: PRIDE, MassIVE
- Metabolomics: MetaboLights, Metabolomics Workbench
- Clinical: ClinicalTrials.gov (for trial data with results)
- Consortium research portals (AD Knowledge Portal, NF-OSI, PsychENCODE, HTAN, DREAM challenges and others): `Synapse_search_entities` finds public Synapse projects/folders/files/tables/datasets by keyword (`node_type` narrows it), `Synapse_get_entity` returns the curated annotations (assay, species, tissue, disease), and `Synapse_list_children` browses a project. Anonymous and read-only: restricted entities return a clear "not public" error and file contents are not downloaded -- data access still goes through synapse.org.

**Layer 3 — Literature-based discovery:**
Many datasets aren't in any repository — they're described in paper methods sections.
- Search PubMed/EuropePMC for papers that analyzed the relationship you're interested in
- Read their methods: "We used data from [DATASET NAME]" tells you exactly what exists
- Check supplementary materials for deposited data (GEO/SRA accession numbers)
- This is often the MOST effective strategy for finding niche datasets

## Step 3: Evaluate Dataset Fitness

For each candidate dataset, assess these dimensions:

**Variables:**
- Does it contain your SPECIFIC exposure and outcome variables?
- Are they measured the way you need? (biomarker vs self-report, continuous vs categorical)
- Are key confounders available? (missing confounders = biased analysis)

**Design match:**
- If you need longitudinal: does it follow the SAME individuals over time? How many waves? What's the follow-up interval?
- Beware: "repeated cross-sections" (different people each wave) are NOT longitudinal
- If you need experimental: is there a proper control group? Randomization?

**Sample:**
- Is the sample large enough for your analysis? (logistic regression needs ~10 events per predictor)
- Does the population match? (age range, geography, clinical characteristics)
- Are there subgroups you need? (stratified by sex, race, disease status)

**Access:**
- Publicly downloadable (best) vs registration required (days) vs collaboration agreement (months) vs restricted (may be impossible)
- Data format: CSV/TSV (easy), XPT/SAS (need conversion), proprietary database (may need special software)

**Quality:**
- Is it from a well-known study with published methods? (NHANES, HRS, UK Biobank = high quality)
- Has it been used in peer-reviewed publications? (indicates data is usable)
- What's the response rate / missingness pattern?

## Step 4: Download and Analyze

Don't stop at finding datasets — download and analyze them. Write and run Python code via Bash. Never describe what you "would do" — execute it.

### Data Loading Cookbook

Choose the loader that matches your data source. When unsure of the format, download a small sample first and inspect.

```python
import requests, io, pandas as pd

# --- Tabular files (most common) ---
df = pd.read_csv("data.csv")                                # CSV / TSV (use sep="\t" for TSV)
df = pd.read_excel("data.xlsx")                              # Excel
df = pd.read_stata("data.dta")                               # Stata
df = pd.read_sas("data.xpt", format="xport")                # SAS transport (XPT)
df = pd.read_sas("data.sas7bdat", format="sas7bdat")        # SAS native
df = pd.read_parquet("data.parquet")                         # Parquet
df = pd.read_json("data.json")                               # JSON (records or columnar)
df = pd.read_fwf("data.dat")                                 # Fixed-width (some legacy surveys)

# --- Download from URL first, then parse ---
resp = requests.get(url, timeout=120)
content = resp.content
# Detect format from URL or content header
if url.endswith(".XPT") or url.endswith(".xpt"):
    df = pd.read_sas(io.BytesIO(content), format="xport")
elif url.endswith(".csv") or url.endswith(".csv.gz"):
    df = pd.read_csv(io.BytesIO(content))
elif url.endswith(".tsv") or url.endswith(".tsv.gz"):
    df = pd.read_csv(io.BytesIO(content), sep="\t")
elif url.endswith(".json"):
    df = pd.read_json(io.BytesIO(content))
else:
    # Try CSV first, then inspect
    df = pd.read_csv(io.BytesIO(content))

# --- REST API pagination (common for GDC, ClinicalTrials.gov, etc.) ---
import json
all_records = []
offset = 0
while True:
    resp = requests.get(f"{api_url}?offset={offset}&limit=100", timeout=30)
    batch = resp.json().get("data", [])
    if not batch:
        break
    all_records.extend(batch)
    offset += len(batch)
df = pd.DataFrame(all_records)
```

### Merge, Clean, Analyze

```python
# Merge multiple files on participant/sample ID
merged = df1.merge(df2, on="id_col", how="inner")

# Filter population
subset = merged[(merged["age"] >= 60) & (merged["age"] <= 80)].copy()

# Handle missing values
missing_pct = subset.isnull().mean() * 100
print("Missing % per variable:\n", missing_pct[missing_pct > 0].sort_values(ascending=False))
subset = subset.dropna(subset=["exposure_var", "outcome_var"])

# Quick regression
import statsmodels.formula.api as smf
model = smf.ols("outcome ~ exposure + age + sex", data=subset).fit()
print(model.summary())

# Visualization
import matplotlib.pyplot as plt
plt.scatter(subset["exposure"], subset["outcome"], alpha=0.3)
plt.xlabel("Exposure"); plt.ylabel("Outcome")
plt.savefig("/tmp/scatter.png", dpi=150, bbox_inches="tight")
```

Always run the code and report actual numbers (β, p-value, CI, N).

## Step 5: Report Honestly

Structure the report as:

1. **Best available dataset** — name, what it contains, access method, key limitation
2. **Analysis results** — actual statistics (β, p-value, CI, N) from running the code
3. **Alternative datasets** — ranked by fitness, with tradeoffs
4. **What CANNOT be answered** — if no dataset matches the study design needed, say so clearly
5. **Recommended next steps** — apply for access to longitudinal data, replicate in other cohorts

**Critical honesty rules:**
- Never claim a dataset answers a temporal question if it's cross-sectional
- Distinguish "data exists but needs registration" from "data doesn't exist"
- Report actual computed statistics, not hypothetical analyses
- State the strongest analysis possible with available data, even if it's weaker than what was asked

## LOOK UP, DON'T GUESS

Never assume a dataset exists — search for it. Never assume access is public — check. Never assume variables are measured the way you need — verify the codebook.

© mims-harvard · Apache-2.0 · mims-harvard/ToolUniverse

Similar skills

All skills for Science and research →

From the same set136

The whole set →
tooluniverse-epidemiological-analysis The tooluniverse-epidemiological-analysis skill takes observational epidemiology from a PECO question to a statistical report, generating Python code for the analysis. CX Codex +1 258installs tooluniverse-ecology-biodiversity tooluniverse-ecology-biodiversity — a skill for ecology and biodiversity research: species identification via GBIF and NCBI, invasive species impact assessment, IUCN conservation status, population dynamics, and food web analysis. CX Codex +1 257installs tooluniverse-immunology tooluniverse-immunology is a Claude Code skill for immunology research, covering antibody-antigen binding, T/B cell repertoires, and epitope mapping via IEDB, SAbDab, and IMGT. CX Codex +1 256installs tooluniverse-metagenomics-analysis tooluniverse-metagenomics-analysis is a Claude Code skill for exploring microbiome studies with MGnify, GTDB, ENA, and research literature. CX Codex +1 256installs tooluniverse-toxicology tooluniverse-toxicology builds a toxicity picture for a drug or chemical: AOP mechanism, FAERS signals, FDA label warnings, CTD associations, EPA CompTox hazard data. CX Codex +1 250installs tooluniverse-drug-mechanism-research The tooluniverse-drug-mechanism-research skill follows a drug from its primary target through pathway changes to clinical outcomes. Use it for pharmacology research, target comparisons, and mechanism reports. CX Codex +1 250installs tooluniverse-regulatory-genomics tooluniverse-regulatory-genomics queries regulatory elements, transcription factor binding motifs, and predicts non-coding variant effects using genomic databases and deep learning models. CX Codex +1 247installs tooluniverse-data-integration-analysis tooluniverse-data-integration-analysis connects statistical results like DEGs and GWAS hits with biological databases. Explains mechanisms behind p-values. CX Codex +1 246installs

Project sponsors

Companies that keep this analytics open