Hard cost limits are moving from alerting to enforcement: Google Cloud blocks new billable usage for a capped service, while AWS pauses a project at its monthly ceiling. The controls constrain runaway-agent exposure, but callers still need to design for an intentional service pause.
Just published this post about how we’re going to need default hard budget caps on pretty much everything https://t.co/L6XEYpF4uZ
Willison turns a familiar operational fear into a concrete design requirement for coding and personal agents: a cap must stop spend, not merely send a warning. The essay compares the implementation paths emerging at AWS and Google Cloud, including their scope and remaining limits, so the full source remains more useful than its headline.
Git 2.56 is here. ⚡ The new git add --resolved stages only paths that are still marked as conflicted, and refuses to stage them if conflict markers remain. One merge-base case dropped from 0.68s to 0.01s. The highlights 👇 https://t.co/CxbNfofYrf
Git 2.56’s `git add --resolved` stages only previously unmerged paths and refuses to stage a file with leftover conflict markers. The release also reports one merge-base traversal case falling from 0.68 seconds to 0.01 seconds. The feature is narrow—it cannot be combined with `-u` or `-A`—but it makes a common merge-failure state explicit before CI.
35 source items · 67% editorial confidence
This is a compact map of the plumbing behind monetary-policy transmission: repos add reserves only while outstanding, reverse repos alter the liability mix, and the standing repo facility caps overnight funding pressure. It also places the mechanics against the Fed’s 2019, 2020 and late-2025 liquidity episodes rather than treating repo usage as a generic risk label.
The IMF’s tracker records more than 1,200 announced measures across 175 economies and shows how the response shifted from broad fiscal shields toward price pass-through and demand management. Its value lies in separating exposure from fiscal capacity, financing conditions, inflation, reserves and administrative ability to target households.
The September survey moves past headline inflation into firm behavior: the share reporting no annual price adjustment fell from 27% before the pandemic to about 13%, while monthly-or-faster changes rose to 29%. The full note also tests the tension beneath that flexibility: 37% found adjustment harder than a year ago and 83% report alternative pricing strategies.
367 source items · 68% editorial confidence
loading form …