We’re building a context-aware code review tool and would love some feedback

[niklasgruener]

Hey everyone — hope this is an okay place to post this.

We’re a couple of interns at Bitmovin in Vienna, currently building Context Goblin, a code review tool that looks beyond the diff and tries to reason about changes in the context of the wider codebase.

While testing it on real open-source changes, we replayed and reviewed PR #2140. The review is visible here: niklasgruener/crawlee-python#1.

It raised a few points around the proxy-client cache, redirect edge-case coverage, and the responsibilities inside _request_with_redirects. After reading the original review, we noticed that several of these overlapped with things your maintainers had already identified, and that the unbounded cache was a deliberate trade-off. That was still useful feedback for us because it showed where the tool found something real but lacked the context of an already-made design decision.

We’d genuinely appreciate any feedback on whether this kind of review is useful, too noisy, or missing important context. If anyone would be open to letting us try it on a future PR, we’d be very happy to do that too.

If this kind of post or unsolicited review isn’t welcome here, just let me know and I won’t bother you with another one. Sorry if I missed any community rules. Honestly, we’re just trying to find our first few users and customers and thought we’d aim high :)