Explore Data
Data & Analyticsby anthropics
3/3 audits passApache-2.0
Profile and explore a dataset to understand its shape, quality, and patterns. Use when encountering a new table or file, checking null rates and column distributions, spotting data quality issues like duplicates or…
Explore Data teaches the agent to profile an unfamiliar table or file before analysing it. It starts with structure: row and column counts, the grain, whether the primary key is unique, and how fresh and how deep the data is. Each column is classified as an identifier, dimension, metric, temporal, text, boolean or structural field. The profile then covers null rates, cardinality and top and bottom values for every column, percentiles and zero or negative counts for metrics, length, whitespace and case checks for strings, and gaps or future dates for timestamps.
It flags quality issues with explicit thresholds (over 5% nulls warns, over 20% alerts), cardinality surprises, placeholder values such as N/A or 999999, duplicate natural keys and skew, then looks for foreign-key candidates, hierarchies and correlations and recommends dimensions, metrics and follow-up analyses. A schema documentation template is included. It works against a connected data warehouse or an uploaded CSV, Excel, Parquet or JSON file, and asks for one of them when neither is available. It profiles and recommends; it does not carry out the full analysis.
In BusinessMCP, a workspace that connects its own database gives the AI business analyst read-only query access, so this playbook can profile real tables before any reporting is built on them.
What you can do with it
- Profile an unfamiliar warehouse table before writing reports on it
- Find high null rates, duplicates and placeholder values in a CSV
- Document a table's grain, keys and update frequency
- Choose which dimensions and metrics are worth slicing by
Run it on your business data
Imported into BusinessMCP, Explore Data becomes a playbook your AI business analyst applies to your web analytics, funnels and connected databases.
Use Explore Data in BusinessMCPInstall it in a coding agent
One command adds Explore Data to your project.
npx skills add https://github.com/anthropics/knowledge-work-plugins --skill explore-dataHow we vetted it
- Source
- anthropics/knowledge-work-plugins at da38ec1
- Licence
- Apache-2.0
- Security audits (skills.sh)
- Gen Agent Trust Hub: Pass · Socket: Pass · Snyk: Pass
- Bundled scripts
- None, instructions only
Checked 2026-09-25 against its skills.sh listing. How we vet skills
Related skills
All skillsData Quality Frameworks
Implement data quality validation with Great Expectations, dbt tests, and data contracts. Use when building data quality pipelines, implementing validation rules, or establishing data contracts.
SQL Queries
Write correct, performant SQL across all major data warehouse dialects (Snowflake, BigQuery, Databricks, PostgreSQL, etc.).
Statistical Analysis
Apply statistical methods including descriptive stats, trend analysis, outlier detection, and hypothesis testing.
Frequently asked questions
What checks does Explore Data run?
Null counts and rates, distinct counts, top and bottom values, numeric percentiles, string length and whitespace checks, date ranges and gaps, and boolean true rates, followed by a quality assessment and relationship discovery.
Which data sources does it work with?
A connected data warehouse, or an uploaded CSV, Excel, Parquet or JSON file. With neither, it asks for a table name or upload and can suggest profiling queries from a described schema.
How do I install Explore Data?
Run `npx skills add anthropics/knowledge-work-plugins --skill explore-data`, or import it from the BusinessMCP dashboard to use it as a playbook in your workspace.
Keep exploring
Run Explore Data against your whole business
Free plan, no credit card.
Use Explore Data in BusinessMCP