languid-photic

5 ms·

31.
by languid-photic 9mo ago
Naively tested a set of agents on this task. Each ran the same spec headlessly in their native harness (one shot). Results: Agent Cycles Time ───────────────────────────────────────────── gpt-5-2
32.
by languid-photic 9mo ago
I manage a maintained fork and periodically rebase our changes on top of upstream. In this case, rebasing is nice because our changes stay in a contiguous block at the top (vs merging which would interleave them), so it's easy for me a
33.
by languid-photic 10mo ago
I think often it's a question of naivety rather than maliciousness. > AI should, from the core be intrinsically and unquestionably on our side That would be great and many people are working to try to make this happen, but it's
34.
by languid-photic 10mo ago
I am happy to know this term now, thanks. I do think this is part of the alignment problem. There are two side, the agent (here I think there was a gap in institutional knowledge about what is and isn’t appropriate) and the environment (wha
35.
by languid-photic 10mo ago
You're right, this is a bit of a conflation. The curl and lockfile examples aren't sandbox escapes, the network blocks worked. The agent just masked the failure or corrupted local state to keep going. The env var leak and director
36.
by languid-photic 10mo ago
You're right that a Podman container with minimal mounts would have blocked the env var leak. Our sandbox uses OS-level policy enforcement (Seatbelt on macOS, bubblewrap on Linux) rather than full container isolation. We’re using a min
37.
by languid-photic 10mo ago
Yes, typically the agent is just trying to do what it's been instructed to do, but sometimes it's too naive to realize its approach is a bit sketchy. And actually, one way we've hardened our sandbox is by tasking agents with
38.
by languid-photic 10mo ago
> Have you had any "learned the hard way" moments? A big lesson for us is that you still need to be careful even in a sandbox. We've been running Claude/Codex/Gemini in sandboxed YOLO mode and have seen some inte
39.
YOLO in the Sandbox(voratiq.com)
2 points by languid-photic 10mo ago|0 comments
40.
Show HN: Pit Claude, Codex, and Gemini against each other, and apply the best(github.com)
2 points by languid-photic 10mo ago|0 comments
41.
by languid-photic 1y ago
Yes, but I think part of the reason for llms.txt is optimize context. eg beyond content, the human docs often have styling markup which wastes tokens.
42.
by languid-photic 1y ago
IIRC Claude is Cursor’s most used model… so this makes sense from a utility perspective. This is not a new opinion, but it doesn’t make a lot of sense from an Apple-as-a-completely-vertically-integrated-company perspective.