2 ms·
Show HN: Security Cards – Reducing insecure AI-generated code by 72%
AI coding agents often generate functionally correct but insecure code. To address this issue, we have open-sourced Security Cards, targeted security guidance for 80+ widely used libraries across 13 programming languages. Our evaluation shows that the Security Cards reduce the rate of insecure code generation by up to 72.3% in Claude Code with Opus 4.7. You can find these security cards on GitHub (https://github.com/Reware-Labs/securitycards https://github.com/Reware-Labs/securitycards) or on our website (https://securitycards.rewarelabs.com/ https://securitycards.rewarelabs.com/)
We are planning to constantly support more languages and libraries, and we’d be glad to hear your feedback, especially on where these cards would be most useful in your workflow, and which libraries we should support next.
- mh0x90 2mo agoThese only cover libraries? no general secure coding best practices?
- hajipour 2mo agoThanks a lot for the comment. We’re planning to add more general security cards for cases where we don’t yet cover a specific library. However, based on our experience, more general rules can sometimes negatively affect functional correctness. That’s why, for now, we’re focusing primarily on library-specific cards.