Wallabag on Kodi: Curated Reading From the Sofa
The earlier reading pipeline had a deliberate shape:
Chosen RSS sources -> phone inbox -> Wallabag -> e-reader
Each part had a job. The phone was useful for discovery. Wallabag was the
private store and reading queue. The e-reader was the quiet place for long,
focused reading.
There is another useful endpoint for that queue: the television.
That may sound like a strange place to read articles. A television is not an
e-ink display, and a Kodi remote is not a keyboard. But a Kodi device is often
the most available calm screen in a home: it is already connected, it is
comfortable to use from the sofa, and it does not begin with an algorithmic
feed.
The goal is not to move every article to the TV. The goal is to add another
intentional place where the reading queue can appear.
+-----------------------+
| Chosen RSS sources |
+-----------+-----------+
|
v
+-----------------------+
| Phone: scan and save |
+-----------+-----------+
|
v
+-----------------------+
| Private Wallabag |
| backend |
+-----+-----------+-----+
| |
v v
+---------------+ +----------------+
| E-reader | | Kodi / TV |
| slow reading | | shared reading |
+---------------+ +----------------+
Wallabag remains the centre of the system; devices are interchangeable
reading endpoints.
Why Kodi Belongs in the Pipeline
The original setup deliberately separated discovery from reading. Kodi extends
that separation to place and posture.
There are situations where an e-reader is ideal: a long essay, a train ride,
or time away from other screens. There are also situations where a television
is better suited:
- Reviewing a saved article with someone else.
- Reading a short analysis, report, or announcement while relaxing at home.
- Looking at an article where diagrams and photographs matter.
- Opening a deliberately selected queue instead of a recommendation-driven
streaming home page.
The important point is that Kodi is not asked to become a browser. A browser
on a TV tends to bring back the web's usual distractions: tabs, search boxes,
news homepages, and the temptation to wander.
Instead, Kodi opens a list created elsewhere and owned by the reader.
RSS decides what enters the inbox.
Wallabag decides what is worth keeping.
Kodi is one place to read the selected result.
The Wallabag Kodi Add-on
This repository contains plugin.wallabag, a Kodi 19+ add-on written for
Python 3 and designed to work on LibreELEC as well as normal Kodi installs.
It connects directly to a Wallabag server using the public OAuth API.
The add-on provides three small actions:
| Action | Purpose |
|---|---|
| All articles | Browse the saved Wallabag queue, newest updates first. |
| Search articles | Find something already stored in the archive. |
| Article reader | Read cleaned article text and cycle through images saved with the article. |
The interface is deliberately modest. There is no social layer, no trending
section, and no attempt to create an endless content surface. A list of saved
articles is enough.
When an article is selected, the add-on fetches its Wallabag content and opens
a dedicated reader window. The window keeps the text scrollable with the
remote and shows article images in a separate area. If an article contains
several images, the remote can move through them without leaving the text.
That means the rendering path is still controlled by Wallabag:
Original web page
|
v
Wallabag fetches and stores a readable copy
|
v
Wallabag API returns saved content and image references
|
v
Kodi add-on renders a remote-friendly reader
The Kodi device does not need a general-purpose browser account, a feed-reader
account, or a copy of every original website. It only needs access to the
Wallabag backend.
Installing on LibreELEC
Kodi add-ons are ordinary ZIP packages. Build the package from this repository:
make package
The Wallabag package is created at:
dist/plugin.wallabag.zip
Copy it to the LibreELEC machine, for example:
scp dist/plugin.wallabag.zip [email protected]:/storage/
Then use Kodi's Add-ons -> Install from zip file flow and select the copied
archive. This is preferable to copying individual source files into Kodi's
add-ons directory because Kodi can register the add-on and its version itself.
After installation, open the add-on information page and select Configure.
The settings dialog asks for:
| Setting | Meaning |
|---|---|
| Wallabag server URL | The base URL, such as https://wallabag.example.org. Do not append /api. |
| OAuth client ID and secret | Credentials created in Wallabag's API-client management page. |
| Wallabag username and password | The account whose reading queue should appear on Kodi. |
| Entries per page | The size of each browse page. |
The credentials are stored in Kodi's local add-on settings. They should not be
put in the repository or inside a ZIP that will be shared with someone else.
A Device Is Not a Reading Strategy
It is easy to mistake a new screen for a new workflow. Installing Wallabag on
Kodi does not automatically make reading more intentional. The useful part is
still the curation before the television is turned on.
If Wallabag receives every link, the Kodi list will become another overwhelming
inbox. The same rules from the phone workflow still help:
- Subscribe to sources deliberately.
- Scan RSS quickly.
- Save only articles worth returning to.
- Archive articles after reading them.
- Keep the unread queue small enough to feel inviting.
Kodi works best when the Wallabag list is a shelf, not a warehouse.
The Value of Multiple Endpoints
Self-hosting is often described only in terms of privacy or control over data.
Those are important, but there is another benefit: the same saved information
can be presented appropriately in different contexts.
| Context | Best endpoint |
|---|---|
| A few spare minutes, away from home | Phone and RSS reader |
| Capturing a worthwhile link | Phone share sheet and Wallabag app |
| Long, private reading | E-reader and KOReader |
| Shared or sofa-based reading | Kodi and the Wallabag add-on |
| Finding an older article | Wallabag search, including through Kodi |
The backend is stable; the surface can change.
That is the opposite of a platform feed. In a platform model, the platform
owns the surface, controls the ranking, and decides which client experience is
available. In this model, the reader owns the archive and can build small
clients around real situations.
Platform model: one platform -> one feed -> one set of incentives
Personal model: one archive -> several readers -> chosen contexts
Kodi is simply another reader in that second model.
A Small Boundary Worth Keeping
There is a temptation to add recommendations, external browsing, automatic
queues, and many more content sources to a TV add-on. Those features can be
useful, but they risk rebuilding the same attention surface this pipeline was
intended to avoid.
The useful boundary is simple:
Kodi should show what I saved, not persuade me to save more.
That keeps the television from becoming another input machine. It becomes an
output device for a queue that was curated with care.
The pipeline now has several endings, but one beginning: sources selected by
the reader. Whether an article is read on a phone, an e-reader, or a television
is a matter of context. The decision about what deserves attention remains
with the person reading.