What file types does Rig read from S3?

CSV, Parquet and JSONL files load as rows, and a PDF loads as one row per page.

How does Rig avoid loading files twice?

It tracks each file's modification date and reads only files that are new or changed. To stop a rewritten file adding its rows again, set a primary key when you connect.

Does it work with Google Cloud Storage?

Google Cloud Storage has its own connector. This one is for S3 buckets.

How often does the data refresh?

Rig Ingest syncs on a schedule you set.

Files & productivity Supported

CSV / Parquet (S3) integration for Rig

The easiest way to connect CSV / Parquet (S3) to Claude and other AI tools

An S3 bucket is Amazon's object store, where systems and partners drop scheduled exports as CSV, Parquet, JSONL and PDF files.

Try Rig Book a demo

What Rig syncs from CSV / Parquet (S3)

Once connected, Rig models these objects so questions resolve to the right tables and joins:

CSV files
Parquet files
JSONL files
PDF pages

How CSV / Parquet (S3) connects to Rig

Rig Ingest reads the bucket with an AWS access key you provide, loads every file matching a pattern into one table you name, and on later syncs reads only files that are new or changed. An optional primary key replaces a row that arrives again.

Read the step-by-step CSV / Parquet (S3) setup guide

Turn a folder of exports into one table

Many systems write a file a day to a bucket and nothing reads them. Rig loads every matching file into one table, so a month of exports becomes a single table you can query and trend.

Ask Rig

Show totals by day across every export in the bucket
Which days have no export file?
How did yesterday compare with the same day last week?

Keep the latest version of each record

Feeds often resend a row when it changes. With a primary key and a latest-version column such as updated_at, Rig keeps one current row per key, so counts do not double.

Ask Rig

How many distinct orders are in the feed?
Which records changed between the last two files?
Show the current status of each order in the export

Join partner and vendor feeds to your data

Partners often deliver data as files in a bucket. Once the files are a table, Rig joins them to your own systems, so a feed can be reconciled with what you already hold.

Ask Rig

Reconcile the partner's settlement file with our orders
Which customers in the feed are missing from our CRM?
Compare the supplier's stock file with our inventory

Ask your CSV / Parquet (S3) data in Claude, ChatGPT or Cursor

Rig serves your CSV / Parquet (S3) data to AI assistants over MCP. Once CSV / Parquet (S3) is synced into your warehouse, you can ask questions in plain English from the tool you already use, and every answer is backed by validated, governed SQL.

Frequently asked questions

What file types does Rig read from S3?: CSV, Parquet and JSONL files load as rows, and a PDF loads as one row per page.
How does Rig avoid loading files twice?: It tracks each file's modification date and reads only files that are new or changed. To stop a rewritten file adding its rows again, set a primary key when you connect.
Does it work with Google Cloud Storage?: Google Cloud Storage has its own connector. This one is for S3 buckets.
How often does the data refresh?: Rig Ingest syncs on a schedule you set.