# How GitHub’s agentic security principles make our AI agents as secure as possible

> Quelle: GitHub Security Advisories — https://github.blog/ai-and-ml/github-copilot/how-githubs-agentic-security-principles-make-our-ai-agents-as-secure-as-possible/

## Maßnahmen

- [ ] Betroffenheit im eigenen Stack prüfen: Versionen/Komponenten abgleichen.
- [ ] Originalquelle / Hersteller-Advisory lesen: https://github.blog/ai-and-ml/github-copilot/how-githubs-agentic-security-principles-make-our-ai-agents-as-secure-as-possible/
- [ ] Verfügbaren Patch oder Workaround einspielen und dokumentieren.