Every's week: map-driven autoresearch, standards-checking word processor, more
every · x · 2026-11-19
Every's weekly roundup of shipped projects:
- Map-driven autoresearch: an agent reads a map of 104 methods and decides which experiments to run; the explainer video is made entirely in code.
- Galley: a word processor that checks each line against editorial standards as you type.
- Fable 5.2 designed a four-channel amplifier end-to-end, from parts selection to schematic and PCB.
- The Frontier team published vibe checks for GPT-6.2 Astra and Sol with a live page.
- Everycut was ported to Windows and Linux in a day, plus an RSS feed of all 43 experiments.
More from coding & agent
- Devs flag GPT's defensive programming for hiding bugs; OpenAI asks for /feedback cases — pvncher · 2026-11-19
- Flutter ships A2UI 1.0: agents respond with real interactive widgets, not chat text — rseroter · 2026-11-19
- Skills vs MCP vs RAG vs Memory: the four context mechanisms AI agents actually need — MaryamMiradi · 2026-11-19
- Anthropic details agent memory: 'dreaming' lets Claude organize its experience — daniel_mac8 · 2026-11-19
- The case for externalized memory as the real fix for continual learning — daniel_mac8 · 2026-11-19
- FT: US tech outages blamed on rush to fix bugs found by AI — nordicinst · 2026-11-19