Anthropic accidentally shipped a source map in their Claude Code npm package, exposing the full CLI source code. Key findings include: an anti-distillation mechanism that injects fake tools into API requests to poison training data scrapers; an 'undercover mode' that hides AI authorship in open source commits with no way to
Table of contents
Anti-distillation: injecting fake tools to poison copycats #Undercover mode: AI that hides it’s AI #Frustration detection via regex (yes, regex) #Native client attestation below the JS runtime #250,000 wasted API calls per day #KAIROS: the unreleased autonomous agent mode #Other notable bits #So what? #Sort: