this Anthropic harness paper f*cking insane
This is a dev post classified by AGK Intelligence as AI dev tools (a tutorial), kept by the Dev Radar because it carries real work, not commentary.
this Anthropic harness paper f*cking insane they studied 132 enterprise repos and 200,000 Claude Code runs and proved why raw agents collapse by turn 14 to control autonomous agents you need the exact verification guardrails that chat interfaces strip away put Opus 5.5 into an unmanaged chat window and it completes 18% of tasks, wrap it in this 9-step harness and completion hits 95% every boundary, every formula, the 4-plane split across Opus, Sonnet, and Haiku 5.5, all 9 equations spelled out 99% of developers approve a 600-line PR because tests pass, then spend 3 days untangling silent
Posted by marfin (5.6k followers) 2 h ago · 28 likes · 1k views · view the original post on X. Kept by the Dev Radar as AI dev tools.
More dev work like this
- “This answer feels better” is a useful instinct when working with AI. But how do you… — @every
- Big update to qwenFS: it can now hold 4x more files (128, up from 32) and persist them… — @TheVixhal
- People kept asking for an Amp Android app. One Friday night, @sqs had the agent build… — @AmpCode
- took 2-3 iterations, but the GIF maker is going to be fun! — @theaaron
- 🚀 Novita Sandbox is now available in Miles by @radixark . — @novita_labs
- 让 Claude Code 装上 Papermorph 这个 Skill 后，给它一本课本 PDF，就能做成带旁白的互动网页课，故事能做成纸偶动画，书还能做成纪录片和竖屏短片。 — @GitHub_Daily
- I use agents from the terminal, but to be fully honest I spend 50% of my time in… — @giuseppegurgone
- What’s new in Lovable, featuring the people who built it! — @Lovable
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.6k posts from 5k X accounts over the last 21 days, 9.4k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-05 15:59 UTC. Full method.