← All plugins
DUCKDBactive

DuckDB

SQL over files: query CSV, Parquet, JSON and Excel in place — on disk, over https, or in S3-compatible object storage (MinIO, Ceph, R2) — with globs read as one table, schema discovery, bucket listing, file-to-table import and query-to-file export. DuckDB is embedded; there is no server.

The repository Full catalog entry Install Inflowenger
SDK
Go
Protocol
inflowv1
Actions
7
Version
v0.1.0
Author
FloMorphic
Runs on
Any inflowv1 host
Categories
Databases and storesFiles and storage

Reach for it when

The data is files — CSV, Parquet, JSON, Excel — on disk, over https, or in a bucket. SQL over them in place.

The use people don't expect

The folder-of-weekly-spreadsheets problem, solved without a database. A glob like s3://bucket/logs/2026-*.csv is one table you can GROUP BY, and there is no server to run — DuckDB is linked into the plugin.

A flow that uses it

Each one is a shape of flow, not a template to copy. Node names are the labels as they appear on the canvas; a name with a slash means either would do.

01

The folder of spreadsheets nobody owns

Starts withMonthly, or whenever someone drops a new file in.

  1. DUCKDB

    Lists the prefix and finds what's new since last time.

  2. DUCKDB

    Describes each file — columns, types, stats — so a changed header is caught, not inherited.

  3. DUCKDB

    Queries the whole glob as one relation. Reads cap at 1000 rows, because every row enters the flow's context.

  4. Contract

    Branches on whether the numbers are within tolerance.

  5. DUCKDB

    Exports the result back to Parquet — the export has no row cap, since that data never passes through the flow.

What you getThe reconciliation that lived in one person's Excel habit becomes a flow anyone can read, and the month it breaks you find out from the describe step rather than from a wrong total.

The reference, not the pitch

This page is about what the plugin is for. Every action, every connection field, the ports, the install one-liner and the caveats are in the catalog entry, which is the thing the maintainer updates.

DuckDB in the catalog How plugin nodes work

Near it in the catalog

CLICKHOUSEClickHouse

Many rows, asked questions repeatedly — events, logs, metrics, anything you'll want a GROUP BY over.

Google Workspace (OpenConnector)Google Workspace (OpenConnector)

Sheets, Docs, Drive and Calendar — thirty per-operation actions, under a centrally connected account.

MSSQLMicrosoft SQL Server

SQL Server or Azure SQL — including the branch instance that legally cannot leave the building.