Permissions set what an LLM agent can do. Policies decide what it should do. Why system prompts aren't security controls, and where each belongs.
Container security, supply chains, and the tooling gaps nobody built yet
I'm a Staff Solutions Architect at Chainguard, where I help teams adopt secure software supply chains. My career spans from early '90s networking through modern DevSecOps, container security, and AI implementations. Break Glass Notes is where I write about the practical gaps I hit in the field — the questions dashboards can't answer, the feeds that don't exist, and the tooling I end up building to fill them. Expect posts on container images, SBOMs, FIPS compliance, CVE tooling, and the occasional deep dive into registry APIs.
Markdown supported · ⌘↵ to post
No responses yet.