Developer experience

Why we built docs you can run

July 8, 2026·5 min read

Most API documentation is a brochure with syntax highlighting. You read it, nod, copy a snippet, paste it into your editor, hunt for your API key, fix the placeholders, run it, and — if you're lucky — see a response a few minutes later. Every one of those steps is a place to give up.

We deleted them. On Ledgerline, every code example is pre-filled with your test key and has a Run button. Press it and the call executes against the same backend the dashboard reads from. The JSON renders inline. A second later, the event shows up in your live feed.

The key is the environment

There's nothing to configure before your first success because the key is the configuration. A sk_test_… key selects the sandbox; a sk_live_… key selects production. Logged in, we inject your real test key into every snippet on the site. Logged out, you get a shared sample key so the examples still run right away.

Reading is integrating

Because the docs and the product share one live backend, there is no gap between "learning" and "building." The invoice you create from the quickstart is a real invoice in your account, with a real hosted page you can pay with 4242 4242 4242 4242. By the time you've finished reading, you've already integrated.

Why it matters more every year

Runnable examples aren't just nice for humans. An AI coding agent that can execute a call and read the response integrates correctly on the first attempt, instead of guessing at parameter names. As more integration work is done by agents, the runnable example becomes the canonical spec — and the funnel.

Try it yourself: make your first call →


Ledgerline is a cshell reference venture. Product behavior described here is real in test mode; no real money moves in this build.

Keep reading