What we know
Claude Code is an AI coding assistant that asks for user approval before running shell commands on a developer’s machine. A report claims this file can inadvertently expose sensitive information like passwords and API keys if it is committed to public GitHub repositories.
Why it matters
The "don’t ask again" feature is designed to improve developer workflow by reducing repeated permission prompts. This raises concerns about security and privacy for developers using Claude Code. The report highlights a potential risk, but no independent verification or broader analysis of the vulnerability has been published.
What is still unknown
No independent tests have confirmed the accuracy of the reported leak or assessed the scope of affected languages and environments. Details about Claude Code’s general release timeline and any planned patches or mitigations are not available. It remains unclear how widespread the issue is or whether the tool’s design inherently risks exposing credentials.