Hi Nicolas, Nicolas Graves <ngra...@ngraves.fr> skribis:
> have a tremendous amount of data (for git commits, git commit messages, > email exchanges) that could be used to train specialized AI agents that > could be very useful in CI / development contexts. A few examples on the > top of my mind: > - a commit message help complying with GNU standards > - a "rebuilder" agent, to try and rebuild a package that has failed when > the change is trivial (such as ignoring a test / try updating a > dependency). (This could be a "crew" of agents checking for > compilability, lints, build failures... with MCP protocoal and each > agent having its specific training data / RAG). I’m skeptical to say the least, as I wrote elsewhere regarding <https://doi.org/10.1109/SC41406.2024.00090> — but this is off-topic. > Even if that would demand a coordinated effort anyway, I think the move > to codeberg would make this more difficult ; Overall, the API that Codeberg and similar services provides makes it easier to grab data about a project, not more difficult, first and foremost because there’s one uniform API to access everything: issues, pull requests, CI/CD reports, repo activity, and much more. This can be seen in “empirical software studies” such as those submitted to the Mining Software Repositories (MSR) conference. None of what’s done in these studies would be possible when dealing with a patchwork of informally-connected tools like the one we have. Ludo’.
