The Claude Code Source Leak: fake tools, frustration regexes, undercover mode, and more | Alex Kim's blog
alex000kim.com · 1,736 words · saved by 1 readers
Anthropic accidentally shipped a source map in their npm package, exposing the full Claude Code source. Here's what I found inside.
Table of Contents Anti-distillation: injecting fake tools to poison copycats Undercover mode: AI that hides its 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 stuff So what? Update: see HN discussions about this post: https://news.ycombinator.com/item?id=47586778 I use Claude Code daily, so when Chaofan Shou noticed earlier today that Anthropic had shipped a .map file alongside their Claude Code npm package, one containing the full, readable source code of the CLI
saved by
related reading
- Building Auto Mode for Open Models — Benjamin Andersonbenanderson.work
- Detecting and preventing distillation attacks \ Anthropicanthropic.com
- Countering misuse of AI: September 2026 / Anthropicanthropic.com
- Best practices for Claude Code - Claude Code Docsanthropic.com
- How Claude Code is built - by Gergely Orosznewsletter.pragmaticengineer.com
- A Guide to Claude Code 2.0 and getting better at using coding agents – sankalp's blogsankalp.bearblog.dev
- How I Use Claude Code | Philipp Spiessspiess.dev
- Claude 4.5 Opus' Soul Document — LessWronglesswrong.com
- Claude Code is My Computer | Peter Steinbergersteipete.me
- Natural Language Autoencoders \ Anthropicanthropic.com
- Investigating three real-world incidents in our cybersecurity evaluations \ Anthropicanthropic.com
- cchistory: Tracking Claude Code System Prompt and Tool Changesmariozechner.at