I think the challenge is that everyone can now build apps
But
1) almost nobody has distribution (like an audience), or
2) the money to pay for distribution (ads or UGC), or
3) the creative genius to get distribution for free (classically called guerilla marketing)
AI writes the code. Humans review it. Without serious code review: bugs up 23%, review time up 12%. With good review: routine tasks cut 46%. We automated writing. Not judgment. Is your review process keeping up? π
@gregisenberg yeah the stuff around the models is moving faster than the models now. a good claude code setup changed way more for me than any single version bump tbh
Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.
Available today at the same price.
Weβve shipped a security-guidance plugin for Claude Code that helps identify and fix vulnerabilities as youβre writing code.
Available for all Claude Code users. Install from the plugin marketplace (/plugins).
Built this with Claude Code in ~1h: a skill to stop juggling analytics dashboards.
PostHog, GA4, Search Console, one per SaaS. Too many tabs.
Now: /saas-metrics. One command, one report across all my projects π
Curious how you guys handle this across multiple projects ?
Every AI agent = broken focus.
Permission? Switch terminal.
Agent waiting? Switch again.
Vibe Island: macOS menu bar app that monitors all your agents live π€―
1 click to approve. Jumps to the right terminal tab instantly, i love it !
how are you guys handling multi agents ?