Analyzes datasets and system behavior to verify hypotheses with concrete evidence.
Install
mkdir -p .claude/skills/research-dmitryprg-ai && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/18234" && unzip -o skill.zip -d .claude/skills/research-dmitryprg-ai && rm skill.zipInstalls to .claude/skills/research-dmitryprg-ai
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Analyze data, investigate datasets, debug with data, and explore system behavior. Use when analyzing, running SQL queries, data profiling, investigating patterns, building statistics, exploring CSV/JSON data, or debugging with data evidence.Key capabilities
- →Verify data accessibility
- →Examine data structure and types
- →Check for data risks like nulls and duplicates
- →Formulate and test hypotheses with queries
- →Document findings with evidence
How it works
The skill follows a structured workflow to load data, analyze its schema, profile for risks, formulate hypotheses, and conduct experiments with focused queries to verify assumptions.
Inputs & outputs
When to use research
- →Debug system with data
- →Profile database schema
- →Analyze CSV/JSON datasets
About this skill
Data Research Protocol
Principle: Data first, code second.
Jumping to conclusions without examining the data leads to building solutions for problems that don't exist. Every assumption needs verification with actual data.
Workflow
- LOAD — Verify data is accessible (DB connection, file exists, API responds)
- SCHEMA — Examine structure, types, relationships before analyzing
- PROFILE — Check for risks: nulls, duplicates, outliers, wrong types
- HYPOTHESIS — Formulate what you expect to find (and what would disprove it)
- EXPERIMENT — Test ONE hypothesis at a time with a focused query
- DOCUMENT — Record findings with evidence, not opinions
Schema Analysis
PostgreSQL
-- Schema inspection
SELECT column_name, data_type, is_nullable
FROM information_schema.columns WHERE table_name = 'target';
-- Data profiling
SELECT count(*), count(DISTINCT column_name),
count(*) FILTER (WHERE column_name IS NULL) as nulls
FROM target;
-- Distribution
SELECT column_name, count(*) FROM target GROUP BY 1 ORDER BY 2 DESC LIMIT 20;
TypeScript
console.log(`Records: ${data.length}`);
console.log(`Keys: ${Object.keys(data[0] || {})}`);
const nullCount = data.filter(item => item.field == null).length;
const uniqueCount = new Set(data.map(item => item.field)).size;
Risk Profiling
| Risk | How to detect |
|---|---|
| Missing data | Count nulls per column |
| Duplicates | Compare total vs distinct count |
| Wrong types | Check actual types vs expected |
| Outliers | Look at min/max, percentiles |
Mini-Experiment Protocol
EXPERIMENT: [Description]
HYPOTHESIS: [What we expect]
METHOD: [Query or code]
RESULT: [Actual output]
STATUS: CONFIRMED / DISPROVED
Rules: one question per experiment, fast execution, logged results, compared with expectations.
Cognitive Bias Prevention
- Analyze ALL data, not just the first N records (survivorship bias)
- Actively look for evidence that DISPROVES your hypothesis (confirmation bias)
- Check edge cases and outliers, not just the median (anchoring bias)
Safety
- Read-only database queries (SELECT only)
- Always include LIMIT on exploratory queries
- Never modify production data during research
When not to use it
- →When the goal is to modify production data
- →When the task does not involve data analysis or debugging with data
Limitations
- →Limited to read-only database queries
- →Requires explicit LIMIT on exploratory queries
- →Cannot modify production data during research
How it compares
This skill applies a data-first protocol for investigation, systematically verifying assumptions with actual data before considering code changes, unlike immediately jumping to code-based solutions.
Compared to similar skills
research side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| research (this skill) | 0 | 4mo | No flags | Intermediate |
| mimic | 0 | 2mo | Review | Advanced |
| embu-data-audit | 0 | 3mo | Review | Intermediate |
| senior-data-engineer | 21 | 7mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
mimic
mkurman
MIMIC (Medical Information Mart for Intensive Care) database toolkit. Curated ICU data: vitals, labs, medications, notes, diagnoses. Tools for querying MIMIC-III/IV, building ML features, and reproducing benchmarks.
embu-data-audit
murzoglu
Audit active EMBU data pipeline decisions in the doktoratezi repository. Use when working on EMBU, EMBU-C, EMBU-P, Likert 4pt/6pt, Stage 1, Stage 2, Stage 3, family matching, outliers, ICC, CFA, mixed Likert families, find_embu_columns, PII removal, or EMBU v2.0 validation.
senior-data-engineer
davila7
World-class data engineering skill for building scalable data pipelines, ETL/ELT systems, and data infrastructure. Expertise in Python, SQL, Spark, Airflow, dbt, Kafka, and modern data stack. Includes data modeling, pipeline orchestration, data quality, and DataOps. Use when designing data architectures, building data pipelines, optimizing data workflows, or implementing data governance.
senior-data-scientist
davila7
World-class data science skill for statistical modeling, experimentation, causal inference, and advanced analytics. Expertise in Python (NumPy, Pandas, Scikit-learn), R, SQL, statistical methods, A/B testing, time series, and business intelligence. Includes experiment design, feature engineering, model evaluation, and stakeholder communication. Use when designing experiments, building predictive models, performing causal analysis, or driving data-driven decisions.
spark-optimization
wshobson
Optimize Apache Spark jobs with partitioning, caching, shuffle optimization, and memory tuning. Use when improving Spark performance, debugging slow jobs, or scaling data processing pipelines.
finance-manager
ailabs-393
Comprehensive personal finance management system for analyzing transaction data, generating insights, creating visualizations, and providing actionable financial recommendations. Use when users need to analyze spending patterns, track budgets, visualize financial data, extract transactions from PDFs, calculate savings rates, identify spending trends, generate financial reports, or receive personalized budget recommendations. Triggers include requests like "analyze my finances", "track my spending", "create a financial report", "extract transactions from PDF", "visualize my budget", "where is my money going", "financial insights", "spending breakdown", or any finance-related analysis tasks.