Clauday.

Framework choice now has a price tag in agent dollars. Nolan Lawson had Claude Code build the same app in 12 JavaScript frameworks, five runs each, and checked every result against…

A team of small open models just matched a frontier model's one-shot score on hard competitive programming, and the write-up is more honest than most papers. Memdoor's team-solve e…

Take the two flagship lab harnesses, Claude Code and Codex, and run them on someone else's model. On one hard real-world task, both delivered nothing. That is the result of kw-ben…

Google launched Gemini 4 Carbon on October 28, and the first people who get to use it are not developers. "Starting this week, we will make Carbon available to a subset of cyber de…

If you build agent harnesses, three papers from Tuesday's arXiv batch are worth an afternoon, because they push back on the instinct to add more structure. The first tests root-ca…

Four papers in Tuesday's arXiv batch land on the same uncomfortable point: checking each piece of an agent in isolation tells you very little about what the assembled agent will do…

Two papers in Monday's arXiv batch look at what agents do when the normal path is blocked, and neither answer is "ask a human." ImpasseBench built twelve realistic environment blo…

Everyone is adding a second model to check the first one. Three papers in Monday's arXiv batch show how easily that checker gets steered by what it is shown. "Who Verifies the Too…

For two years agent evals scored the model and treated the harness as plumbing. Three papers in Monday's arXiv batch score the plumbing, and the plumbing turns out to decide a lot.…

Two papers in Monday's arXiv batch measured what happens to a coding agent's context when it gets compacted, and they agree on the uncomfortable part. The summary keeps what you as…