Obsidian is the strongest argument in this category and it is not close. Your notes are Markdown files in a folder you chose. There is no account, no sign-up at all, and the free plan has no limits on anything. If the app disappeared tomorrow your notes would be exactly where you left them, readable in TextEdit.
That is the standard every local-first app should be held to, including mine. So the honest framing of this post is narrow: there is one reason people leave Obsidian, it is not a feature, and it does not apply to everyone.
The one honest reason
You wanted a notes app. You got a system to maintain.
Obsidian ships deliberately spare and expects you to assemble the rest. That is a real design philosophy and for a large number of people it is the entire appeal. The vault becomes theirs in a way a finished app never can be. Community plugins, themes, CSS snippets, a daily-note template, a task syntax, a folder convention, a plugin that adds a calendar, another that adds kanban.
Then, some months in, a plugin breaks against an update. Or you sit down to write something and spend forty minutes on the vault instead. Or you realise you cannot remember which of the eleven plugins is responsible for the behaviour you rely on.
None of that is Obsidian being bad. It is the cost side of extensibility, and it is invisible when you are enjoying it and obvious when you are not. Linking notes without building a second brain is about the same trap from the other direction: the structure starts as a tool and becomes the work.
Three things people get wrong about the price
Obsidian is free, and it is worth being precise about what that does and does not cover. All figures from obsidian.md/pricing, checked 1 September 2026.
Sync is a subscription. $4 per user per month billed annually, $5 monthly. Files-on-disk means you can sync them yourself with iCloud, Dropbox or Syncthing, and plenty of people do, but the supported, encrypted, conflict-aware option is a recurring bill, and self-syncing a folder of Markdown across two machines is its own small project.
Commercial use is $50 per user per year. This is the one almost nobody knows. If you use Obsidian for work at a company of two or more people, the commercial licence applies. It is not enforced by the app and there is no phone-home, which is why it stays invisible, but "free" describes personal use, and a lot of Obsidian's heaviest users are using it at work.
Catalyst is $25 once, and buys early access and a badge rather than features you need. Genuinely optional.
So the price comparison is rarely $0 against anything. For a personal, single-machine, hand-synced vault it really is free, and that case is common. For someone syncing two machines and using it at work, it is $98 a year.
What Obsidian does that a finished app cannot
Say this plainly, because it decides most of these choices.
Files you can read anywhere. Markdown in a folder is the best exit story in software. Nothing beats it, including a database — getting your notes out is trivial by construction.
Linking and graph. Backlinks, unlinked mentions, transclusion. If you write by connecting ideas rather than filing them, this is the feature and there is no adequate substitute.
Plugins. Whatever workflow you have, someone has built for it. This is unbeatable by any single-developer app, permanently.
Every platform. Mac, Windows, Linux, iOS, Android.
If any two of those matter to you, stop reading. The rest of this is not aimed at you and switching would cost you things you cannot get back.
The case for leaving anyway
It is narrow and it is about scope, not quality.
Obsidian is a notes app. Tasks, a calendar, habits and a clipboard history are plugins, or other apps, or both. That is fine when the plugins hold. The alternative shape is an app where those things are simply present, built by one person against one data model, so there is nothing to install and nothing to break in October.
You trade extensibility for not having to think about it. That is a bad trade for a tinkerer and a good one for someone who wants to open the app and start working. Only you know which you are, and the honest test is not what you would like to be. It is whether you enjoyed the last hour you spent configuring the vault.
Moving out
The export is the part that is already done, which is the nicest thing about leaving Obsidian.
Your vault is the export. A folder of .md files. Copy it somewhere safe before you touch anything.
Wiki links do not travel. [[Note name]] is Obsidian's convention, not Markdown's. Whatever you import into will render those as literal text unless it understands them. Budget for that if your vault is heavily interlinked, and consider it a signal. A collection whose value is mostly in its links is a collection that should probably stay in Obsidian.
Plugin-generated content is Markdown, mostly. Dataview queries are not: they are code that produces a view at render time, and they arrive as unresolved query blocks. Anything a plugin was computing for you needs rebuilding by hand or letting go.
Attachments live in the vault. Check where your plugin config put them before you copy folders around.
The honest version
Cyanote does not compete with Obsidian on Obsidian's own ground and would lose if it tried. There are no plugins, no graph view, no backlinks panel, no mobile app, no Windows build on sale, and your notes are in a SQLite database rather than a folder of files — exportable, but not editable in place.
What it is: notes, to-dos, a board, a calendar, habits, routines and a system-wide clipboard in one Mac app, finished, with nothing to install, for $4.99 once and no account. If the vault has become a hobby you did not sign up for, that is the trade on offer. If the vault is the point, stay. You are already using the best thing there is.