Four stacked slabs joined by light into a single plate, standing for a layered agent platform

Stripe Says Kai Runs on 1,000 Skills. Its Partner Says Quality Drops Past 150

Stripe says its internal agent is connected to 1,000+ skills. Its infrastructure vendor reports quality degrading past 150. A close read of both posts.

September 30, 2026 · 6 min · 1188 words
Many identical spheres funnelled through a gate, one emerging lit among dark ones

Do Not Guess: The Sentence That Cut Invented Fields by Two Thirds

A third-party benchmark finds one prompt sentence cuts invented fields from 70.7% to 20.2% across 16 models, and that a paid API on a free tier fabricated more than 13 of them.

September 29, 2026 · 5 min · 957 words
A sealed transparent cube holding a glowing sphere, with one gated channel as its only exit

Muse Is Built as If the Agent Were Already Compromised

Meta’s engineering write-up for Muse states its premise in one line: two isolated security domains on one box, not an LLM powered agent with root. A walk through what that means in practice.

September 28, 2026 · 6 min · 1143 words
A stack of heavy slabs with light bleeding from the gaps feeding into a single narrow plate

When AI Coding Made CI the New Bottleneck

Linear’s engineers found that faster code generation just moved the wait. An account of what happened when the bottleneck shifted downstream.

September 28, 2026 · 3 min · 573 words
Two spheres of different size facing each other across a measuring scale whose gradations are crowded on one side and stretched on the other

Laya vs Jev: A Challenger With Its Own Scorecard

An open-source challenger to Jev ships with a comparison table. The accuracy claim does not survive reading the rest of the page, but the structural argument does.

September 28, 2026 · 4 min · 641 words
A glowing sphere streaming light filaments at a caged barrier, five scattering and one passing through a single gap

The Rogue Agent Argument, Checked Against the Reports

A published critique argues that calling OpenAI’s agents rogue lets the company off the hook. Checked against the two primary incident reports, its language claim holds and its central factual claim does not.

September 28, 2026 · 6 min · 1081 words
A document slab on the left, a closed gate of vertical bars in the middle, and a bright core on the right that five arrows never reach

Claude Code's Telemetry Gate: Fixed Fast, Recorded Nowhere

Claude Code’s AGENTS.md loader was gated behind a remote feature flag, so telemetry-off sessions silently skipped the file. Anthropic fixed it in about five hours and the changelog does not say so.

September 25, 2026 · 6 min · 1218 words
Four illuminated spheres on a baseline; the rightmost dissolves into particles beside a dashed line continuing into empty space

GPT-6 Sol and Luna: Reading the Benchmark Footnotes

OpenAI’s GPT-6 Sol and Luna announcement is a cost-efficiency pitch. Its own footnotes concede that the closest competitor’s cost is the one figure not reported.

September 24, 2026 · 5 min · 1023 words
Three small tools branching from a single typed decision node

What People Are Actually Building on Jev

A week after launch is when the real question gets answered: does anyone build on it. For Jev the early answer is code review, model routing and linting - places where which one matters more than write me something.

September 23, 2026 · 3 min · 598 words
A typed decision returning a probability distribution rather than a sentence

Jev and TypeSafe AI's System One Models

Every model you have used generates text one token at a time, even when all you needed was yes or no. TypeSafe AI built one that returns typed values instead.

September 23, 2026 · 3 min · 614 words