OpenCLI: Turn Any Website into a CLI

The pitch is almost too simple: point OpenCLI at a website, get a CLI for it. No API, no official SDK — it figures out the structure and wraps it in a command-line interface you can script against.

My first reaction was skepticism. Web scraping-based tools have a reputation for being fragile, and for good reason. But the use case here is narrower and more honest than it sounds.

Who it's actually for

Not for production pipelines — you wouldn't want to depend on this for anything critical, since a site redesign could break your commands overnight. But for personal automation against sites that don't expose an API? This is genuinely useful.

Think: internal tools at companies that have web UIs but no REST endpoints. Legacy software. Sites where you're the only user and you just want to script some repetitive action without setting up a full browser automation rig.

The interesting part

What caught my attention isn't the scraping — it's the interface generation. Getting the mapping from "page structure" to "sensible CLI flags and subcommands" right is the hard problem, and from what I can tell they've thought carefully about it.

It's the kind of tool that feels unnecessary until you hit exactly the problem it solves, at which point you're glad it exists.