How much time do you waste trying to run a new GitHub repo?

4 points by prabhavsanga 6mo ago ↗ HN

I'm building a tool where you paste a GitHub URL, and it auto-detects the stack (Node, Python, Go, etc.), spins up a sandbox, and runs it immediately. No config, no git clone, no dependency hell.

My use case is auditing open-source tools or testing libraries quickly without polluting my local machine or dealing with version conflicts.

My question to the community:

Do you use existing tools (Codespaces, Gitpod) for this? I find them too "heavy" for just checking if a repo works.

Is the friction of npm install / pip install high enough that you'd use a "one-click run" service?

Or do you prefer reading the code statically and never running it?

1 comment

[ 0.22 ms ] story [ 11.8 ms ] thread

palata 6mo ago ↗

I personally choose not to depend on more wrappers. If I need to clone a git repo, I `git clone` it. Then I build and run the project using the build system of the project.

If the project properly uses a build system I am familiar with, then I don't really need to think. If the project does something exotic, then chances are that I will just give up on that project. But I don't think that your tool would help here: if it is exotic, then your tool probably won't know how to automate it.