Is Claude 4 a snitch? I made a benchmark to figure it out

Links

• soydev.link/firecrawl
• soydev.link/chat
• simonwillison.net/2025/May/31
• www-cdn.anthropic.com/6be99a52cb68eb70eb9572b4cafad13df32ed9
• snitchbench.t3.gg
• github.com/t3dotgg/snitchbench
• soydev.link/sponsor-me
• t3.gg