According to Apiiro security research, Anthropic's Claude Code exposed passwords, API keys, and credentials in 1,122 public GitHub files through its 'Don't Ask Again' feature, which allowed developers to disable permission prompts without understanding the security implications. The State of AI reported Goldman Sachs subsequently restricted Claude Code and GitHub to engineers only, cutting off traders and bankers who had used the tools for monitoring client flows and building market models.
Topics #
Sources #
- Press[Read article](https://apiiro.com/blog/coding-agents-leaking-secrets-github-dont-ask-again)
- Press[Read article](https://www.thestateofai.com/news/goldman-sachs-claude-code-traders-bankers)
Go deeper #
This intelligence is sourced automatically from public sources across the web and synthesised by the Prefactor AI pipeline. Stories are reviewed before publication.
source & further reading
getreadyforagents.com — original article
AJX launches open-source CLI and skill bundle for testing product usability from AI agent perspective
Plannotator releases inbox interface for human review and annotation of AI agent work outputs
Ars Technica: Chainalysis hacked via multifront attack including AI agents; all customer credentials stolen