Data Analytics Skills
31 skills for Claude Code and Codex for the analyst workflow: data checks, cohorts, funnels, A/B tests, dashboards, summaries
- 529 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- The skills are instructions and templates, run nothing themselves and connect to no databases
- Following the instructions the agent may run queries on data you open to it, and the permissions come from your agent settings
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/data-analytics-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skills are instructions and templates, run nothing themselves and connect to no databases
- Following the instructions the agent may run queries on data you open to it, and the permissions come from your agent settings
This is third-party code. Review the repository files before installing.
What it does
The repository holds 31 portable skills for Claude Code and Codex in six categories, from data checks to analysis retrospectives. Each skill is a folder with a SKILL.md, sometimes with scripts and references. The agent activates the right skill from your task description: there are no slash commands and no schemas or metrics to configure in advance. A skill asks for the minimum context, runs a step-by-step process, names its assumptions explicitly and returns the result in a uniform template. The first category checks data: programmatic-eda profiles a dataset, data-quality-audit assesses quality against business rules and schema, and query-validation reviews SQL. Documentation and knowledge: semantic-model-builder, metric-reconciliation, schema-mapper and more. Analysis: cohorts, segmentation, funnels, time series, root-cause-investigation for a metric jump and ab-test-analysis for experiments. Data output covers insight-synthesis, visualization-builder, executive-summary-generator and dashboard-specification. The rest is stakeholder communication and workflow organization: analysis-planning, peer-review-template and analysis-retrospective.
Who it is for. Data and product analysts working in Claude Code or Codex who want to structure routine analysis steps without company-specific setup.
Good fit when
- Before working with a new table: programmatic-eda checks structure, nulls, outliers and correlations
- The activation metric fell 12% in a week and root-cause-investigation walks through hypothesis testing
- Experiment results need an analysis with significance through ab-test-analysis
- Findings have to be packaged for leadership through executive-summary-generator and insight-synthesis
Not a fit when
- The skills do not connect to databases themselves: you bring the data and context and the agent runs queries with your own means
- You need scripts and notebooks for one company: the set is deliberately generic
- You need a visualization tool: visualization-builder prepares chart recommendations and specs but does not draw them
Example request
Our activation rate dropped 12% last week. Investigate the cause and lay out hypotheses step by stepLimitations
- Installation is manual: clone the repository and copy the folders you want into ~/.claude/skills for Claude Code or ~/.agents/skills for Codex, and the README has no installer command.
- The skills are generic and ask context questions before working, so without your information about metrics and data the result stays general.
- A skill works with what you give it: if data is missing it states assumptions and proceeds with the work.
- The set has 31 skills and few are needed at once, so copy only the ones you need.
How to disable. Delete the skill's folder from ~/.claude/skills or ~/.agents/skills and restart the agent session. The cloned repository can be deleted entirely and does not affect projects.
FAQ
How do I install Data Analytics Skills in Claude Code?
Clone the repository and copy the folders you want, for example cp -r 01-data-quality-validation/programmatic-eda ~/.claude/skills/. Project skills live in .claude/skills/ inside the project. For Codex the user folder is ~/.agents/skills/.
Which categories are in the set?
Six: data quality and validation, documentation and knowledge, analysis and investigation, data storytelling and visualization, stakeholder communication, and workflow optimization. 31 skills in total.
Which skill should I start with?
For new data start with programmatic-eda, then data-quality-audit. For SQL use query-validation and schema-mapper, for a metric jump root-cause-investigation, and for a complex task analysis-planning first.
Do I need to configure schemas and metrics in advance?
No. The skills work on demand: each asks for the minimum context, runs its process and names its assumptions. If you did not give all the data, the agent says what it assumed and continues.
Does the set work in Codex?
Yes, the authors cover both agents. For Codex copy skills into ~/.agents/skills/ or the repository root's .agents/skills/, and for Claude Code into ~/.claude/skills/ or the project's .claude/skills/.
Discussion
No questions or reviews yet.
Related
SkillsA template for data scientists: AGENTS.md, CLAUDE.md and skills for Python, SQL, Polars and model diagnostics.
16 Claude Code skills for data engineers: source profiling, failure investigation, change review and documentation as web pages.
Around 200 skills for Claude Code on bioinformatics, chemistry and biostatistics, from RNA-seq analysis to molecular docking
High-Stakes Analytics Lab
High-Stakes Analytics & Decision Lab
An evidence-gated analytics skill for high-stakes decisions: profiles data, picks a method the evidence supports, can answer with a data request
Skill and CLI for tables: cleaning with recorded decisions, level-based statistics, a PDF report and a separate audit.
A set of commands, skills and subagents for Claude Code that runs data analysis like a careful team analyst.
README in short
The README explains how the set differs from approaches that need upfront setup and shows a skill map. Tables for six categories, installation by copying folders, advice on the first skill to pick and a description of the skill workflow pattern follow.
README badge
Original SKILL.md
--- name: programmatic-eda description: Systematic exploratory data analysis. Activate when a dataset needs profiling — structure check, nulls, outliers, distributions, correlations — before deeper analysis begins. --- # When to use - You receive a new dataset and need to understand its shape and quality before analysis - An analysis produces surprising numbers and you want to verify the underlying data first - A stakeholder asks "is this data reliable?" or "what's in this table?" - You're about to run a model or statistical test and need data-quality assurance # Process 1. **Load and overview** — run `scripts/data_overview.py` to get row count, dtypes, memory usage, and a sample. Confirm grain (what one row represents). 2. **Null profile** — run `scripts/null_profiler.py`; compare output against thresholds in `references/quality_thresholds.md` and flag columns above limits. 3. **Outlier detection** — run `scripts/outlier_detector.py` (IQR + z-score) on numeric columns; document flagged values and decide: real signal or data error? 4. **Distribution summary** — run `scripts/distribution_summary.py` for descriptive stats and univariate histograms on each numeric column. 5. **Correlation exploration** — run `scripts/correlation_explorer.py`; flag pairs with |r| > 0.8 as potential multicollinearity or redundancy. 6. **EDA checklist sign-off** — work through `references/eda_checklist.md` and confirm each item before declaring the dataset profiled. 7. **Write findings** — fill `assets/eda_report_template.md` with full profiling output; distil top issues into `assets/findings_summary.md`. For pattern recipes (e.g. polars vs pandas equivalents, chunked reads for large files), see `references/pandas_polars_recipes.md`. # Inputs the skill needs - Required: dataset path (CSV / Parquet / Excel) or a DataFrame already in scope - Required: business context — what does one row represent? - Optional: quality threshold overrides (defaults in `references/quality_thresholds.md`) - Optional: columns to skip (PII, binary blobs, high-cardinality IDs) # Output - `assets/eda_report_template.md` (filled) — full profiling report with per-column stats - `assets/findings_summary.md` (filled) — top 3–5 quality issues and recommended next steps - Console output / plots from scripts for interactive inspection