Loading…
Loading…
The AI tooling landscape changed more in the last 12 months than in the five years before it. New model families, new agent frameworks, new IDE integrations, and a constant churn of point solutions — most of which disappear before they build traction.
This is the pillar page for everything WOWHOW publishes about AI tools: reviews of what actually ships work, tutorials that skip the fluff, and honest comparisons of the frameworks developers are betting on. Start at the top with the latest, or jump to a sub-topic below.
macOS 27 preinstalls fm, a command-line front end to the same on-device and Private Cloud Compute models that run Apple Intelligence. No account, no API key, no per-token bill. fm chat for a conversation, fm respond for scripts, structured JSON output and an OpenAI-compatible local server. This is the shell playbook: five pipe recipes worth keeping, where it fails, and when to reach for mlx_lm or Ollama instead.
The fastest-growing agent repositories in September 2026 are not multi-agent frameworks. They are skill libraries: anthropics/skills, awesome-agent-skills with more than a thousand entries, NVIDIA/skills, 380-skill community packs. A skill is reusable competence a single agent loads on demand. An agent crew is rented attention that talks to itself. This is the decision table, the skill-file anatomy and the daily ritual that turn a strong single agent into a Creator OS.
Two things landed on the Hacker News front page in the same week: Bryan Cantrill’s essay on how readers instantly spot LLM-authored prose, and an arXiv paper by Solé, Krakauer, Levin and colleagues modelling LLM adoption as a contagion with tipping points into dependence. Read together they make a practical argument: draft yourself, then use the model as the editor it is good at being. Here is the tells checklist and the workflow.