Create an owner-approved DESIGN.md for Claude Code
Write a visual specification from approved brand decisions, then use it for two small Claude Code edits and inspect both diffs and renders.
Explore by topic
AI Models & Tools
Explore BIG CHANGE reporting on AI Models & Tools: what changed, how it works and why it matters.
Latest reporting
226 stories
Check Claude Code instruction scope across two client projects
Compare the instructions loaded in two fresh Claude Code project sessions, then verify when a nested instruction file joins context after an authorized read.
Use a pinned DESIGN.md reference in Claude Code for one page
Choose one pinned DESIGN.md file, give it to Claude Code for an edit to an owned page, then inspect the changed code and desktop and phone rendering.
Build a local Realtime voice agent with read-only tool calls
Build a live browser voice conversation with OpenAI Realtime and two synthetic read-only tools. Complete server, browser and controller code shows call ID matching, safe continuation, interruption and microphone cleanup.
How to choose or reject an AI tool for one small-business task
Compare an actual no-code AI candidate with your existing route for one small-business task. Check fit, permissions, full cost and accepted work before choosing or rejecting it.
Build a local browser voice conversation with OpenAI Realtime
Build a local browser microphone conversation with OpenAI Realtime over WebRTC. The full server and page code keep the API key on the server and show how to check playback, stop the microphone and handle failures.
How to rewrite a selected Word passage with Claude and review changes
Select one passage in Word, ask Claude for a bounded rewrite in tracked changes mode, then inspect and decide each revision before saving.
How to inspect context and continue a Claude Code session
Inspect a Claude Code session's context and loaded instructions, compact when needed, then check the task state against real files and acceptance checks before continuing.
Create and check a complete scene in Google Flow
A documented route from permitted Google Flow clips through version choice and Scenebuilder assembly to a downloaded scene checked from beginning to end.
Ask ChatGPT about one open Chrome tab and check its source
Use ChatGPT's Chrome extension to ask one question about a mentioned open tab, then check the answer against the page title, URL and supporting passage.
Copy a Genspark Design prototype and inspect its source ZIP
Duplicate a Genspark Design prototype, verify access to the copy, make one limited edit and inspect its source ZIP using the documented controls.
Harness-Delta Attribution separates agent gains from benchmark shortcuts
A new research method separates detected benchmark overfitting and extra inference compute from the remaining gain in optimized agent harnesses.
FLARE tests AI grid restoration decisions under degraded reports
A new simulation benchmark finds that AI agents often match a grid-restoration reference plan, but their worst cases and unsafe switch proposals keep independent safety checks central.
OpenCode reports breach affecting 1.4% of Console users
OpenCode says a breach affected 1.4% of Console users. Its founder separated hosted Console provider keys from local OpenCode credentials; affected owners should check their provider accounts.
Create, correct and export a local AI slide deck
Set up Agent-Native Slides locally, build three slides from your own factual text, make one exact correction, and check it in a downloaded HTML deck.
Check one Agent-Native action through chat and a React page
Call the Chat template's hello action from agent chat and a React route, change one return value, and compare the documented results while checking for tool invocation evidence.
HybridAgent tests latent reasoning alongside text-based tool use
A new preprint and released code show how HybridAgent alternates latent reasoning with text tool use. Its measured gains come from the authors' benchmark setup.
Liquid’s d1-ultra adds a hosted model for decisions without generated text
Liquid AI released d1-ultra as a hosted model that returns probabilities for structured text and image decisions without generating answer tokens. Its benchmarks and price come from the company.