Claude Code Team Says Review Now Runs "Continuously, More Like Spellcheck While You Write"
Simon Willison quotes the team's post on how Claude reviews and fixes most changes before a human looks.
Simon Willison quotes the team's post on how Claude reviews and fixes most changes before a human looks.
It complements CodeQL for PHP, Shell, Terraform and Dockerfiles; findings are advisory.
A merge ruleset and an endpoint let agents check packages against org license policy.
It can open PRs and issues across repos and use shared Slack threads as context.
The AINews recap for Oct 15-16 also launches a "High Taste Testers" lunch series.
A ChatGPT plugin helped drive up to 10x growth in agent creation.
An OpenAI customer story on longer generated video with voice.
A No Film School podcast with Brian Duffield and Daniel Kraus on adapting the novel.
NVIDIA RTX Spark comes to Photoshop, Premiere and Substance 3D.