推文

@ktwu01 · 2026-10-11 14:31

Big news!!! Our repo keeps getting flooded by AI-generated PRs (worst case: one account, 58 PRs in 10 days). I think I finally found a fix: Anyone trying to DDoS us with PRs now gets DDoSed by their own agent first. It’s simple. Our CONTRIBUTING.md now tells AI agents to stop and ask their human four questions before writing a PR: 1. What outcome do you really want? 2. What’s the highest-leverage variable behind it? 3. Will this work move that variable, or just keep you busy? 4. Who is the customer, and what do they really want? The agent can’t answer for them. It has to wait. Anyone trying to DDoS us with PRs now gets DDoSed by their own agent first. Brilliant!!! Inspired by @ycaophysics.

曝光 58 · 评论 2 · 点赞 0 · 书签 0 · 曝光/时 3.9916686247279665