How it works today

No integrations yet.
No magic.
Here's exactly what happens.

This is the first version, so it's worth being plain about it. Nothing here is automatic yet. Here's precisely what you do, and what Patch does for you, right now.

1

Log what happened, yourself

Write down what you did and why, right when you're doing it. There's no Figma or GitHub integration yet, so this part is manual. It's the one habit the rest of Patch depends on.

2

Watch your timeline build itself

Every capture lines up automatically on your project's timeline, in order, as you go. You don't organize anything. That part already works today.

3

Bring your own key to generate

To turn captures into a written case study, connect your own OpenAI or Anthropic API key in Settings. It takes about two minutes, and it's the only way generation runs in this first version. You stay in control of exactly which model runs.

Required for generation right now
4

Export whenever you want

Everything you've logged, projects and captures both, exports as markdown, images, PDF, or HTML. One click, whenever you want it.

Manual today, more automatic soon. We're building the integrations that log this for you, so step 1 stops being something you have to remember.

See what that looks like on the roadmap.