VersoWrite.·NoughtCalculate.·SkeinConnect.

Blog·12 August 2026

Everything in the box

What 25 million plugin downloads tell you to build - and why none of it is a plugin in Skein.

Skein has no plugin system, and that is a decision rather than a backlog item. It creates an obvious problem: how do you know what to build, when you cannot let other people build it for you?

There is a good answer, and it was sitting in public the whole time. Obsidian’s community plugin directory publishes install counts. That list is not a list of features to copy. It is a ranked list of what people needed badly enough to go and find something, which is a much better signal than any survey I could run.

Read the numbers carefully

The counts are cumulative downloads, not users: first installs plus every update anyone ever pulled. Old plugins are inflated by years of updates, fast-releasing ones by their cadence, and abandoned ones never decay. So the ranking is worth an order of magnitude, not a decimal place. With that caveat, the top of the list is unambiguous.

What people installDownloadsIn Skein 1.1
Dataview, Tasks, Kanban, Calendar13.9MLive queries, four views
Excalidraw7.1MNo - the canvas is cards, not sketches
Templater, QuickAdd6.6MTemplates and quick capture
Advanced Tables3.1MTable editor
Git, Remotely Save4.9MNo - version history, but no sync
Admonition946kCallouts
Tag Wrangler990kRename a tag everywhere
Importer1.45MImport from Notion

The most useful thing in the data

Look at what is near the top and ask what it is really saying. Style Settings and Minimal Theme Settings have 4.1 million downloads between them. Those plugins exist because appearance in Obsidian is CSS with no settings screen. Four million people did not want a plugin. They wanted a font size.

Tag Wrangler is 990k downloads because you cannot rename a tag. Homepage and Recent Files are 2.3 million between them, and each is about a day of work. Advanced Tables is 3.1 million because the editor makes you line up pipe characters by hand.

A large part of that directory is not extensions. It is missing settings screens and unfinished features, wearing the costume of an ecosystem. Which is the clearest argument I have found for the way Skein is built: every one of those is cheaper, better and more coherent done once in core than done by three competing plugins with different keyboard shortcuts.

What one engine buys you

The four database-shaped plugins - Dataview, Tasks, Kanban, Calendar - are four of the top eight, and they are one need wearing four hats: give my notes structure I can query, and let that structure be time and status. Table, list, board and calendar are four renderings of one ordered result.

So in Skein they are one feature. A block like this:

```skein
from: Sources
where: type = source and year > 1950
sort: year desc
as: table
columns: title, author, year
```

Change as: table to as: board and you have a Kanban. Change it to as: calendar and you have the calendar. There is no separate plugin to learn, no second query syntax, and no chance of two of them disagreeing about what a tag is.

What I deliberately did not copy

Dataview has a JavaScript escape hatch. Templater is JavaScript. Both are genuinely powerful, and both turn notes into a place where code runs. Skein’s query language has no functions and no parentheses; its templates substitute a fixed list of tokens and evaluate nothing.

That is a real cost. Some things you can do in Dataview you cannot do in Skein and will not be able to. The trade is that a Skein vault is a folder of text that cannot execute anything, which is worth more to me than the last ten per cent of expressiveness.

The same reasoning kept AI out. Not because the features are bad - local embeddings for “related notes” would suit the graph Skein already draws - but because the app currently makes no network connections at all, and “none” is a much easier promise to verify than “only these ones”.

The honest list of what is missing

Excalidraw is the single most-installed plugin in the ecosystem, 7.1 million downloads, and Skein has no drawing at all. Sync is 4.9 million and Skein has none. Those are the two largest holes and I am not going to pretend otherwise; the canvas holds cards and arrows, and your vault syncs only as well as the folder you keep it in.

There is also no mobile app, and if a feature is not on the features page you cannot add it yourself. That is the actual price of no plugins, and for some people it will be too high. It should be.

Why this is the better trade anyway

A plugin ecosystem moves the cost rather than removing it. Someone still has to find the plugin, trust its author, keep it updated, and work out which of the three that claim to do the same thing is the one people actually use. Then it breaks on a major version, and the notes that depended on its syntax are now notes with syntax in them that nothing renders.

Everything in Skein renders in Skein, and everything Skein writes is Markdown that renders elsewhere. Its canvases are JSON Canvas, the same open format Obsidian writes, so those files open in either app. If Skein stops being maintained tomorrow, you are left with a folder of ordinary files - which is the only promise a notes app can make that is worth anything.

See everything in the box, or download it and have a look. It is free, it is 10.4 MB, and it does not ask for an account.