Anthropic accidentally publishes Claude Code source code for anyone to find

THE DECODER / 4/1/2026

📰 NewsDeveloper Stack & InfrastructureSignals & Early TrendsTools & Practical Usage

共有:

Key Points

Anthropic has accidentally exposed parts of the source code for its AI coding tool, Claude Code, after earlier internal blog post leaks tied to its new Mythos AI model.
The article frames the incident as another security/controls failure, with the disclosed code now reportedly available for anyone to find.
The leak specifically affects an AI development product, potentially enabling easier replication, scrutiny, or modification of parts of Claude Code’s implementation.
The report implies heightened reputational and operational risk for Anthropic as these publication mishaps compound.

Anthropic can't catch a break: Following the recent leak of internal blog posts on its new Mythos AI model, parts of the source code for its AI coding tool Claude Code have now been made public.

The article Anthropic accidentally publishes Claude Code source code for anyone to find appeared first on The Decoder.