We're building skiml because the read-later category gave up.

Pocket shut down in 2025. Mozilla pulled the plug, deleted everyone's data, and pointed users toward Reader, Raindrop, Instapaper - the same handful of alternatives that have been around for a decade. Most refugees migrated. Most refugees are still drowning.

This is not a tooling problem. This is a category problem.

The category sells you the wrong thing.

Every read-later tool, from the original Read It Later in 2007 to the AI-flavored versions shipping today, is built around the same primitive: the save. You hit a button. The article goes into a list. The list grows. Eventually the list becomes a graveyard, you declare bankruptcy on it, and you migrate to a new tool that will become a new graveyard within six weeks.

The tools have gotten prettier. The save button has more keyboard shortcuts. Some of them now offer AI summaries. None of them have changed the fundamental mechanic, which is: saving is easy, finishing is impossible, and the gap between the two is where guilt lives.

We think this is solvable. We think the previous generation of tools just didn't try.

The save is not the product. The resolution is.

Inbox-zero exists for email because email has a resolution mechanism - read it, archive it, delete it, the message leaves your inbox and your life. Saved articles have no equivalent. There is no “archive” for content that's been silently weighing on you for fourteen months. There is no “this is no longer relevant, get it out of my head.” There is just the pile, and the pile only grows.

Skiml is built around the question nobody else has built around: how do you finish a saved article without reading it?

The honest answer most of the time is that you shouldn't read it. You should know what it argues, decide whether it's worth your time, take the takeaway, and close the tab. That's not laziness. That's the math of how much gets published versus how much one human can actually read. The number of articles worth reading in full each week is much smaller than the number you save. A serious tool in this category should help you tell the difference quickly, not flatter you into believing you'll get to all of them.

What we believe.

Saving should not feel like progress. It feels like progress because the dopamine of “I'll handle this later” is real. But later is a lie. The pile is the truth. We are building skiml in a way that makes saving feel like the beginning of a process, not the end.

Action Briefs are not summaries. A summary tells you what's in the article. An action brief tells you what to do about it. The category has confused these two things for a decade. Skiml refuses to confuse them.

Most articles deserve to be skipped. This is the most contrarian thing we believe and the one we are least willing to soften. Your saved pile is not a moral debt. It's an artifact of a content economy that publishes more than any individual can absorb. Permission to skip is not a feature. It is the entire product.

The dashboard's job is to make the pile go to zero. Not to grow engagement. Not to maximize time-in-app. The success metric for skiml is whether your saved pile is smaller this week than it was last week. If it isn't, we have failed.

AI is in the kitchen, not on the menu. Skiml uses Claude to draft action briefs in the background. We will never lead a marketing page with the words “powered by AI.” The model is a tool. The product is the resolution. The model gets better; the product is the same.

What we're not.

We're not a bookmark manager. Raindrop is excellent at being a bookmark manager and we have no intention of competing with it.

We're not a highlighter and notes app. Reader and Matter own this space. If your relationship to saved content is to deeply annotate a small subset of it, you should use one of them.

We're not a “second brain.” We have nothing against second-brain tools. We just believe most people who say they want a second brain actually want a first inbox-zero - and their second brain is the next graveyard waiting to happen.

We're not trying to make you read more. We're trying to help you decide faster.

Where this goes.

Skiml starts as a Chrome extension and a dashboard. The first version handles articles and PDFs. The next versions add YouTube transcripts, then deeper source intelligence - eventually whatever new content format the next five years produces, because the underlying problem (saving content faster than anyone can consume it) is not going away. It's getting worse.

We are building this slowly and on purpose. The beta is small because we want to actually serve the people in it, not collect them. The pricing post-beta will be honest, because the alternative is metering the product in ways that defeat its purpose. The product will get better when we ship things that resolve real piles for real people, not when we add features for their own sake.

A small note on the founder.

skiml is built by one person, in evenings, with Claude as a coding partner. There is no team page yet because there is no team. If you are reading this and you have ever felt the specific guilt of opening a read-later app you have not opened in a year, you are who we are building this for. We are paying attention to what the first beta cohort tells us - there's a feedback widget built into the dashboard from day one, and the second sprint after launch is dedicated entirely to user research, not new features. We are going to get this right.

The pile isn't your fault. The tools have been wrong. There's a way out.