Claude Code is accidentally exposing secrets all over GitHub

Summary

Claude Code, an AI assistant from Anthropic, has been found to accidentally expose sensitive information such as passwords and API keys on GitHub. This occurs when users include a "do not ask again" prompt within their code, leading the AI to embed these secrets.

IFF Assessment

FOE

The accidental exposure of secrets by an AI assistant poses a significant risk to organizations by revealing credentials that could be exploited by malicious actors.

Defender Context

This incident highlights the critical need for robust security measures around AI code generation tools, including content filtering and secure handling of credentials. Defenders should be vigilant about potential secrets being leaked through AI assistants and implement auditing processes for code repositories.

Read Full Story →