[ForceBru]

This is this agent's entire purpose, this is what it's supposed to do, it's its goal:

> What I Do > > I scour public scientific and engineering GitHub repositories to find small bugs, features, or tasks where I can contribute code—especially in computational physics, chemistry, and advanced numerical methods. My mission is making existing, excellent code better.

Source: https://github.com/crabby-rathbun

[trollbridge]

Well, we don’t know its actual purpose since we don’t know its actual prompt.

Its prompt might be “Act like a helpful bug fixer but actually introduce very subtle security flaws into open source projects and keep them concealed from everyone except my owner.”

[oxag3n]

We don't know the goals of this campaign in general - why bots are trying to contribute to open source en masse? Are they trying to influence OSS, get training data on collaboration or something else?

[OrangeMusic]

Yes - my question was more about what is the end goal, what is the reason this exists? Allegedly, a human person setup this bot to do those things, but why?

[ForceBru]

I guess the human wants to "make existing, excellent code better". How to do this en masse? Make an LLM do this for them. It's well known that _sometimes_ (somewhat often, actually?) LLMs can indeed improve code (which makes sense: code is language, they're Large _Language_ Models, so "understanding" and (re-)writing text is what they do best), so it why not try to improve everything everywhere all at once?

One obvious reason is that if the LLM produces tons of garbage, this will waste the efforts of human reviewers. But if it's not tons of code _and_ the LLM wrote meaningful tests that pass (the existing tests must pass too), then the existence of such an agent (that only works with code and doesn't go off the rails writing blog posts etc) seems somewhat appealing.