Design Tokens Only Constrain Agents When the Compiler Can Say No
Sharing one token source between Next.js and Expo is easy. Making it the only path an agent can take on both platforms is the part that matters, and the web side still has a hole.
Sharing one token source between Next.js and Expo is easy. Making it the only path an agent can take on both platforms is the part that matters, and the web side still has a hole.
Claude Fable 5.2's launch was read as a new model at a lower price. Its own charts show something more useful. On my test, Opus 5.5 at max effort scored no better than at medium and cost about 10 times as much. Your team needs an effort policy before it needs a model upgrade.
Coding agents raised code output by about 30% without moving finished work, because review absorbed the gain. Sprint planning still prices the writing. Price the review instead.
Web developers moving to React Native tend to port the metrics and leave the useful part behind. What made Core Web Vitals work was a contract of field data, a fixed percentile and a pass mark set outside the diff. Agents writing your app's code make that contract more important.
Last week's trojanized Claude Code extension never touched a prompt, a permission or Anthropic's build. It came through a mirror marketplace on auto-update. The agent's install channel is now the softest part of your AI tooling, and nobody on most teams owns it.
An approved pull request proves someone read the diff. It doesn't prove anyone can debug it when it breaks in production. For code on a paged path, that second test should decide whether it merges.
In a React Native app, risk depends on which side of the native line a change lands, not on how big the diff is. When agents write the diff, that line has to be checked by a machine before a reviewer ever opens the PR.
The AGENTS.md hijack found in a Downloads folder last week isn't a Downloads problem. Instruction files are code that runs with whatever permissions you've already granted, and a pull request can plant one as easily as a browser can.
When commits per engineer can rise fivefold in six months, a career ladder built on delivery scope stops telling you who is senior. Promote on the judgment you can audit instead.
I measured eight common re-render causes with and without React Compiler, then gave the survivors to an agent. The compiler handles what's in the component. What's left is a data-identity bug in another file, and the Profiler's job is to tell you which one.