AI Blocks › Build and ship software
DuckDB big file query helper
Point it at large CSV, Parquet or JSON files. Get fast queries run from MATLAB with DuckDB.
DuckDB big file query helper is a free AI skill for the job "Build and ship software". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · BSD 2 Clause licence · Made by MathWorks
Made by MathWorks. See the original on GitHub; we copied it as it is, with its licence.
What it does
Large CSV, Parquet and JSON files queried in place from MATLAB, with no giant loading step. You get SQL answers fast, and only the reduced result lands in memory. The skill uses DuckDB through the Database Toolbox. On load it audits your code for the slow habit of reading a file into MATLAB, writing it to a database, then querying it. It sends you straight to reading the file with SQL instead. It covers in-memory and file databases, filtering, grouping, deduplicating, cleaning and sampling at the source, and saving to a .duckdb file for repeat use. It also says when not to use it: small files, or number-heavy work better done in MATLAB afterwards.
You give it
- Paths to CSV, Parquet or JSON files
- The question or summary you need
- Existing MATLAB code to review (optional)
- MATLAB with Database Toolbox
You get
- SQL queries that run on the files directly
- A reduced table ready for MATLAB
- A review of slow loading patterns
- A saved .duckdb file for repeat use (optional)
Use it when
- Your file is too big for readtable.
- Your code loads a CSV only to write it to a database.
- You need to filter and group a large Parquet file before analysis.
- You want to profile a dataset before deciding what to load.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run MATLAB R2026a or later with Database Toolbox.