# Sikkerhed for AI‑agenter

> Udvalgte resultater, der demonstrerer denne ydelse.

- [Har skrevet 1.000 RBAC‑regler til Linux‑servere](https://engineer.company/da/portfolio/wrote-1000-rbac-rules-to-secure-linux-servers/) — ai-security
- [Har ledet adgangsstyring med M365 og SSO](https://engineer.company/da/portfolio/led-access-management-across-m365-and-sso/) — ai-security
- [Har bygget fail‑closed rate limiting mod misbrug](https://engineer.company/da/portfolio/built-fail-closed-rate-limiting-and-anti-abuse/) — ai-security
- [Har kørt et løbende sikkerhedsscanningsprogram](https://engineer.company/da/portfolio/ran-a-continuous-security-scanning-programme/) — ai-security
- [Tre låse på døren, ingen af dem låst](https://engineer.company/da/portfolio/three-locks-on-the-door-none-of-them-locked/) — ai-security
- [Har bevist, at blokeringen virker ved hver kørsel](https://engineer.company/da/portfolio/proved-the-intrusion-ban-works-every-run/) — ai-security
- [Har tilføjet et preflight‑tjek af hemmeligheder](https://engineer.company/da/portfolio/added-a-one-second-preflight-for-secrets/) — ai-security
- [Har skrevet en afgrænsningsregel efter et læk](https://engineer.company/da/portfolio/wrote-a-scope-rule-after-a-credential-leak/) — ai-security
- [Har adskilt alle driftshemmeligheder pr. vært](https://engineer.company/da/portfolio/split-every-operational-secret-per-host/) — ai-security

<https://engineer.company/da/services/ai-security/>