video1.0<iframe src="https://www.loom.com/embed/34ad585695784680b1770dfaad41c525" frameborder="0" width="1732" height="1299" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12991732Loomhttps://www.loom.com12991732https://cdn.loom.com/sessions/thumbnails/34ad585695784680b1770dfaad41c525-afd00fdc6615e02f.gif286.024This Loom demonstrates how chaining two light LLM vulnerabilities can escalate from authenticated low-privilege access to remote code execution. It shows starting with a minted token, failing to run RCE due to lacking admin access, then using a “PE” against an endpoint to obtain a proxy admin role and write an ID to template lmrc, confirming code execution. In Meego, it highlights visibility including the Kubernetes ingresses and services, the deployment, and DeepTrace context with the command and full stack trace. When mitigation controls are enabled, sensor blocking prevents sub-process creation, and a WAF rule returns a 400 for escalation-like requests, resulting in a 403 with a custom error message on reattempts.