Today's launches all narrow what an agent is allowed to be, a named caller or a two-megabyte task instead of a general capability, while the day's biggest story is a lab pointing the same attribution machinery outward at people.

Today's launches all narrow what an agent is allowed to be, a named caller or a two-megabyte task instead of a general capability, while the day's biggest story is a lab pointing the same attribution machinery outward at people.

The most useful numbers published in the last 24 hours are the ones that name where a thing stops working, and the people publishing them are the ones who gain least from saying so.

Ten thousand agents can now be pointed at one problem, and the only thing that makes their output checkable is a formal certificate rather than the fleet that produced it.

The industry stopped arguing about whether agents work and started publishing what they cost, in dollars per researcher per day, in context tokens per skill, and in the gap between cheap findings and expensive fixes.

What an agent loads has become the thing worth managing, and the week's launches are almost all knobs on that inventory rather than new capability.

Last week's shipping was almost entirely about approval gates, machinery deciding what an agent may read and what it may finalize, and GitHub handed an agent the approval bit in the middle of it.

OpenAI spent Sunday publishing its own evidence that the layer watching AI work is falling behind the layer doing it, and two independent pieces from the same week describe the identical failure at human scale.

Agent capability work has moved from the model to the box the model runs in, and this week showed both halves of that shift at once, labs industrializing the manufacture of training environments while the sandboxes already in production kept failing to hold.

Tgrep's headline speedup measures how slow the filesystem is rather than how good the index is, and the durable win for coding agents is…

Quantization damage is nonlinear rather than gradual, so the only defensible way to choose a quant is a task benchmark run against the file…

Per-caller credential resolution only becomes practical when a missing grant pauses the run and asks instead of throwing, which turns a…

The valuable result in deltafin's Kimi K3 run is not one token per second, it is the measurement showing that six-minute prefill is a…

A skill that never fires is usually a description problem rather than a useless skill, so the right response to a cheap unused skill is to…

Ripwire earns trust not with its 52x headline but by re-running its own head-to-head, publishing a corrected margin of 1.46x instead of the…

Verification instructions in a system prompt only change outcomes when they move the agent off a specific default behavior, and describing…

Whether an agent repo is safe to run is decided by its credential surface, its reversibility path, its maintenance recency and its…