Setup & configuration

On first run, Read Flow writes a read-flow.toml file for you — most people never need to edit it by hand, since every setting also has a control in the app’s Preferences UI.

read-flow.toml is specific to your machine: it holds your server’s hashed credentials and local file paths, so it’s never checked into version control and each installation gets its own copy.

Configuration sections

Section	What it controls
[database]	Path to the SQLite database file.
[server]	Bind address/port, allowed origins, and authorized users.
[scan]	Which folders to scan, file types, concurrency, auto-tag rules.
[ui]	Private mode and which tags count as private.
[online_library]	OPDS catalogs to search (Project Gutenberg, Standard Ebooks, …).

Adding a folder to scan

From the app: Preferences → Scan → Add folder. The equivalent config entry looks like:

[scan.directories."/home/you/Documents"]
action = "Scan"
tags = []
inherit = false

• action — "Scan" to index files found here, or "Ignore" to skip a subfolder inside an already-scanned tree.
• tags — tags automatically applied to everything found in this folder.
• inherit — whether subfolders inherit this entry’s tags/action by default.

Running headless

On a home server or NAS, skip the GUI entirely:

read-flow --headless --address 0.0.0.0 --port 8000

See Headless server + PWA for the full walkthrough, and Multi-instance & remote sources if you’re connecting more than one Read Flow server from the same web app.