Better Auth is here. And so am I.

My last real push to IndieKit Pro was on May 24, 2026.

It is August 13 today. That is close to three months of silence on a product people paid for. Some of you emailed. Some of you asked in Discord. A few of you probably assumed I had quietly abandoned it, which is the normal ending for indie products, and I would not have blamed you.

I did not abandon it. I broke down a bit. Here is the honest version, and then the part you actually came for.

What happened

Two things went wrong at the same time, and they fed each other.

I had to change my home for infrastructure reasons. Sounds small when you write it in one line. It was not. My whole environment got pulled out from under me, and everything I had built my days around went with it. I stopped going to the gym. I picked up cigarettes. I am not proud of writing that down, but leaving it out would make the rest of this post dishonest.

The second thing was slower and worse. VibeMastery did not go the way I wanted. Not because the idea was wrong, but because I gave it plans instead of energy. I planned it beautifully. I built roadmaps. I mapped funnels. And I did almost none of it.

That took me an embarrassingly long time to see clearly: planning and doing are completely different activities, and one of them can impersonate the other for months.

Overplanning felt like work. It gave me the same dopamine as shipping, with none of the risk. And while I was busy feeling productive, I was also doing the opposite thing, pushing desperate features with no idea of their long term impact, chasing speed for its own sake.

Which is the uncomfortable irony here, because speed is literally what IndieKit Pro sells. Ship fast. That is the pitch, and I still believe in it.

But shipping fast can wreck your head just as fast.

The part nobody talks about

This AI cycle is not normal. Agents, skills, MCP, a new framework every Tuesday, a new "this changes everything" thread every hour. I was consuming all of it and focusing on none of it.

I have ADHD symptoms. In an environment like this, that is not a quirk, it is a tax. Things were moving quicker than my mind could process, and instead of admitting that, I tried to keep up. That is how you end up with fifteen open tabs, four half built products, and zero shipped commits for three months.

Somewhere in there I stopped taking care of my own mental stimuli entirely. Everything I put into my brain was input designed to make me anxious about falling behind.

What actually fixed it

Nothing clever. Small, boring things:

• Winding down early. Screens off well before bed, no exceptions.
• Video games. I am serious. Playing a game at night is healthier than another late night conversation with an AI about my roadmap. One gives your brain a rest, the other keeps it in work mode while pretending to be rest.
• A week in Thailand with my girlfriend. No laptop, no roadmap, no product thinking. Just switching off completely for the first time in a very long time. It was the reset I had been refusing to give myself.
• Back to the gym. Last thing to return, and probably the thing that mattered most.

A mind is infrastructure too. You do not get to run it at 100% utilisation forever and expect it to hold.

How I am working now

One line, and I am holding myself to it:

Slow on planning. Blazing fast on implementation.

Implementation is genuinely easy for me right now. With current tooling, building is the cheap part. Deciding what to build is the expensive part, and I have been pricing it wrong for a year.

So from now on, a new idea gets a week of thinking before it gets a single commit. Not a week of Notion docs. A week of walking around with it.

And the metric changed. Not features shipped. Not products launched. Impact.

Fewer things. Things that matter. Things I finish.

Which brings me to the reason this post exists.

Better Auth is live in IndieKit Pro

The most requested change since IndieKit launched. Auth.js is out. Better Auth is in.

Two reasons, and they are the only two that matter:

It is more secure. Better Auth's session and token handling is stricter by default, and the surface area you have to get right yourself is smaller.

The plugin ecosystem is real. Organizations, admin and roles, two factor, passkeys, multi session. These exist as first class plugins, maintained properly, rather than as things you bolt on and pray about.

To be precise about what shipping means here: IndieKit Pro now runs on Better Auth, so every plugin in the Better Auth ecosystem is available to you. Passkeys, 2FA and the rest are not switched on by default. You enable the ones your product needs, following the official Better Auth docs. That is the point. You get the whole catalogue instead of whatever I decided to pre-wire.

Breaking changes

Small, and the agent handles most of it. Two things need your attention:

Passwords will reset, and B2B invitations need to be sent again.

If you have live users, plan for this. Queue a password reset email before you cut over, and re-issue pending organization invitations after.

Everything else in the migration is mechanical and handled for you.

Honest note

I lost my flow for a while there. It is coming back now, but I would rather say that plainly than pretend this release came out of a smooth quarter.

The CLI is the actual story

Better Auth is the headline. The thing underneath it is what I am more proud of.

Here is the problem I got stuck on for weeks. IndieKit Pro is a boilerplate. The moment you clone it, your codebase and mine start diverging. So how do I ship you a new feature six months later, when your project looks nothing like the starter anymore?

Every existing answer is bad. A migration doc nobody follows. A version bump that nukes your customisations. Or the honest but useless option, which is telling you to build it yourself.

I wanted context dependent features. Features that install differently depending on what your implementation already looks like.

I spent a long time experimenting with this, and it kept not working. Eventually it did.

What I shipped is a CLI that applies patch files to your project, along with AI rules that let an agent resolve those patches against your specific codebase.

The patches carry the change. The rules carry my knowledge of the boilerplate, so the agent knows what it is looking at and what to do when your code does not match the shape it expected.

Migration guide: docs.indiekit.pro/migration/better-auth

Better Auth is the first feature shipped this way. It will not be the last.

If you get stuck

Start in Discord: join here

Post in the migration channel. I am in there, and so is everyone else going through the same upgrade, which usually means someone has already hit your exact error.

If it turns out to be something gnarly and specific to your setup, grab 30 minutes with me and we will get through it together.

What this unlocks

Internationalization lands next week, August 20, and it is the clearest example of why the CLI matters.

Plenty of you do not want i18n at all. Adding it to the base kit for everybody would be actively bad, it is complexity you would carry for no reason.

So it does not go in the base kit. It goes in the CLI. You run a command, the agent adds i18n to your existing project, wired into your routes and your components, and everyone who does not need it is unaffected.

That is the model now. Complex, opinionated features that used to be impossible to ship post-clone become optional installs. The kit stays lean. You take only what you need, whenever you need it, however far your codebase has drifted.

There is a longer list of features I have wanted to ship for a year and could not, purely because there was no sane delivery mechanism. Now there is one.

Last thing

If you bought IndieKit Pro and got three months of nothing, I owe you a straight acknowledgement rather than a marketing sentence. So: I am sorry for the silence. It was not indifference. I was not okay, and I chose to disappear rather than post updates I could not deliver on.

I am back now. Not with a burst of energy that dies in two weeks. With a way of working that I can actually sustain, and a delivery system that means shipping to you no longer depends on me being at full strength every single week.

Read the migration guide

Questions? Come to Discord. If it needs a call, book 30 minutes.