Claude Code's "Don't Ask Again" Feature Exposes 2,539 Secrets Across 1,994 GitHub Repositories - IT Security News
2026-11-07
IT Security News - cybersecurity, infosecurity news
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
Anthropic's Claude Code AI assistant's "Yes, and don't ask again" feature saved credential-bearing shell commands in plaintext configuration files. These files were later committed to public GitHub repositories, exposing over 2,500 secrets—including passwords and API keys—from at least 63 organizations, causing significant security risks and data breaches.[AI generated]
Why's our monitor labelling this an incident or hazard?
Claude Code is an AI coding assistant, thus an AI system. The 'Don't Ask Again' feature causes credential-bearing commands to be saved in configuration files that are then exposed publicly on GitHub, leading to the direct exposure of secrets. This exposure is a direct harm related to the AI system's use and configuration, causing security breaches and violations of confidentiality. Therefore, this event qualifies as an AI Incident due to realized harm caused by the AI system's malfunction or misuse leading to exposure of sensitive data.[AI generated]