Connect SQLite to Excel, Sheets and AI

Reading a file that something else is using

SQLite's locking is what stops most reporting tools dead: the application that owns the file holds it, and the report either blocks or fails. Immutable mode is the way out, and it is a deliberate choice rather than a hidden default.

• Immutable mode opens the file with no locking at all, so a live application keeps working and your query still returns.
• It is opt-in on purpose. It tells SQLite the file will not change while it is being read, which is the right promise for a report and the wrong one for a file being actively rewritten by a long transaction.
• Left off, the file is still opened strictly read-only — the driver itself refuses to write, on top of the statement check that runs before anything is sent.
• Nothing is copied. There is no temporary snapshot, no upload, and no second version of your data to keep track of.
• The journal mode of the file is reported back on the connection test, so you can see whether it is running in WAL before deciding.

An encrypted file is fine

• An encrypted SQLite database opens with the passphrase you supply. It is held on the agent, like every other credential here.
• Views are listed alongside tables, so a file that keeps its reporting logic in views is browsable as it was designed to be.
• Indexes are read too, which is how the Query Builder knows what will be fast.

One file, or a folder of them

This connector is for a single file. If what you have is a folder of SQLite files with the same shape — one per store, one per day, one per customer — the SQLite Folder connector unions them into shared tables and tells you which file each row came from.

-- local_db is the connection; SQLite's only schema is main
SELECT   o.status, COUNT(*) AS orders
FROM     local_db.main.orders      o
JOIN     pg_crm.public.customers   c ON c.id = o.customer_id
GROUP BY o.status