Read File

This is a practical wrapper around DuckDB that lets Claude read pretty much any data format you throw at it: CSV, JSON, Parquet, Excel, shapefiles, SQLite databases, even Jupyter notebooks. It handles local files and remote URLs from S3, HTTPS, GCS, and Azure. The implementation uses a single macro that pattern matches file extensions and routes to the right DuckDB reader, then shows you the schema, row count, and a 20 row preview. Useful when you want Claude to quickly inspect or answer questions about a dataset without writing format specific code yourself. It will auto install DuckDB if missing and retry with the right extensions if spatial or Excel files fail on first attempt.

skills.sh installs: 1.3K
GitHub stars: 616
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add duckdb/duckdb-skills --skill read-file --agent claude-code

Installs into .claude/skills of the current project.

Files

You are helping the user read and analyze a data file using DuckDB.

Filename given: $0 Question: ${1:-describe the data}

Step 1 — Read it

RESOLVED_PATH is $0. If the user gave a bare filename (no /), resolve it to a full path with find first.

Run a single DuckDB command that defines the read_any macro inline and reads the file.

For remote files, prepend the necessary LOAD/SECRET before the macro:

Protocol	Prepend
`https://` / `http://`	`LOAD httpfs;`
`s3://`	`LOAD httpfs; CREATE SECRET (TYPE S3, PROVIDER credential_chain);`
`gs://` / `gcs://`	`LOAD httpfs; CREATE SECRET (TYPE GCS, PROVIDER credential_chain);`
`az://` / `azure://` / `abfss://`	`LOAD httpfs; LOAD azure; CREATE SECRET (TYPE AZURE, PROVIDER credential_chain);`