---

name: kw-data description: Use when write SQL, explore schemas, build visualizations and dashboards, and validate analyses across any data warehouse or CSV upload.

Data (Query, Visualize, Interpret)

Anthropic's Data Analyst plugin turns Claude into a data collaborator. With a warehouse connection (Snowflake, Databricks, BigQuery, or any SQL-compatible database via MCP), Claude queries directly, explores schema and table metadata, runs end-to-end analyses, and iterates on queries based on results. Without a connection, paste SQL results or upload CSV/Excel files for analysis and visualization — Claude will also write SQL for you to run manually and then analyze the results you provide. Slash commands cover query writing, dashboarding, and validation.

What you get

• Direct warehouse querying via MCP (Snowflake, Databricks, BigQuery, SQL-compatible)
• schema and metadata exploration
• visualization and dashboard building
• CSV/Excel upload path when no warehouse is connected
• SQL written for manual execution + result analysis
• connectors Snowflake, Databricks, BigQuery, Definite, Hex, Amplitude, Jira

Customize your output

• Connector mode: warehouse (Snowflake/Databricks/BigQuery) vs standalone CSV/Excel
• SQL dialect: any SQL-compatible
• Output: query, chart, dashboard, narrative insight

Example output

Given a Snowflake connection and 'why did MRR dip in June', Claude explores billing and subscription schemas, writes the churn cohort query, charts new vs churned MRR by plan, and flags a mid-June price-test cohort as the driver.

Best for

Data analysts, analytics engineers, and PMs running SQL and dashboards on a warehouse or spreadsheets

---

Source: The 12 Cowork Knowledge-Work Skills · https://github.com/anthropics/knowledge-work-plugins/tree/main/data Category: Data / Analytics · Priority: P3