I’ve been using Beancount for about five years now, and I recently tried to build a reconciliation plugin that would save me hours each month. Seemed straightforward in theory—hook into the import process, add some custom logic. But I hit a wall pretty quickly. The AST documentation is sparse, examples are few, and honestly, I wasn’t sure if what I built would even get adopted. It made me wonder: should we be thinking differently about how we support community contributors? I’d love to hear from others who’ve attempted this. Are the barriers worth it, or should we lower them?
tried building one last year. got lost in the AST, docs weren’t helpful, gave up. not worth the time investment tbh.
Here’s something I’ve been thinking about—if we assume 100 people in the community have plugin ideas but only two or three actually succeed in shipping something, that’s an enormous amount of lost value. The barrier isn’t technical skill; it’s friction. Most of those people have the capability but lack accessible documentation, a clear registry where plugins live, and mentorship from experienced contributors. If we invested in those three things—even a basic wiki, a plugin directory, and a mentorship matching system—we might unlock 10x more contributions. That’s the compounding benefit of lowering barriers.