Yeah that’s what I’m experimenting with, but I think it’s overengineered, especially with the whole dogmatic SPARC approach. I’m personally a more minimalistic person, and I would prefer it to be natively integrated into the app and being able to define exactly the (system) prompts for each of the agents.
> Slack, a Salesforce-owned workplace messaging app, recently blocked other software firms from searching or storing Slack messages, The Information reported on Tuesday, citing a public disclosure.
Yea the proxying/observability is without question the simplest part of this whole problem space. Once you get into the weeds of automating all the eval and prompt optimizing, you realize how irrelevant wireshark actually is in the feedback loop.
But I also like you landed on mitmproxy as well, after starting with tcpdump/wireshark. I recently started building a tiny streaming textual gradient based optimizer (similar to what adalflow is doing) by parsing the mitmproxy outputs in realtime. Having a turnkey solution for this sort of thing will definitely be valuable at least in the near to mid term.
Looks very buttoned up. My local project has some features tuned for my explicit agent flows however (built directly into my inference engine), so can't really jump ship just yet.
Later in the thread it seems like https://github.com/mricon/b4 was involved. Maybe just a bug this time but exposes it as a weak link in the whole kernel contributor web of trust
I refuse to be nerd sniped; do you know what the input to the SGI is and what it outputs? looking at the video it seems that most of that is done "in hardware", the SGI could just be providing the actual updated information, and it could just be for nostalgia or "if it is not broke..."
reply