DuckDB with Obsidian

Obsidian-duckdb-motherduck is a plugin for Obsidian that works similarly to Obsidian Dataview, but with SQL.

It can run arbitrary data with SQL, as DuckDB supports almost any format, you could query local CSVs, XML, or anything else. But not only local files work, but also remote files via DuckDB’s Extensions such as HTTP(S), Huggin Face with hf: or anything else.

How it looks inside Obsidian:

Cool thing, you can freeze the output, to show up and keep in your notes, so no need to run again. But you can also set it to auto refresh.

This is also really nicely implemented, as it’s just plain markdown added back to the note. Meaning if I publish this on my Public Second Brain with Quartz, it will show as well, as seen below examples.

Query remote parquet file - and free in-line:

1
2
3
4
5
6
7

SELECT
  o_orderpriority AS priority,
  count(*) AS orders,
  round(sum(o_totalprice), 2) AS revenue
FROM read_parquet('https://shell.duckdb.org/data/tpch/0_01/parquet/orders.parquet')
GROUP BY 1
ORDER BY revenue DESC

Top authors on Bluesky:

1
2
3
4
5

SELECT author, COUNT(*) as post_count 
FROM bsky.posts 
GROUP BY author 
ORDER BY post_count DESC
LIMIT 10;

• I also wrote about Building an Obsidian RAG with DuckDB and MotherDuck

Origin: GitHub - motherduckdb/obsidian-duckdb-motherduck: Obsidian Plugin for DuckDB & MotherDuck · GitHub