← 🤖 AI Infrastructure
Advanced
Prompt injection in repository automation: defend the execution boundary
Threat-model AI automation reading pull requests and repositories. Separate untrusted text from execution authority and test approvals against changed artifacts.
View Premium accesscourse lessons, concepts and answers · see current terms
RELATED CONCEPTS
PRACTICE THIS IN REAL QUESTIONS
Infrastructure as Code & ConfigurationResources keep changing outside Terraform and plans show surprises weeks later. How do you detect drift early, and who may change things?→Linux, Networking & ScriptingExplain what actually happens when cron fires a job, and where its silent failures come from.→Infrastructure as Code & ConfigurationA major provider release lands and forty stacks pin the old version. How do you roll the upgrade forward without a fortnight of fire drills?→AI & GPU InfrastructureYour team wants an AI agent that can open pull requests and trigger deploys. What is the risk model?→AI & GPU InfrastructureA coding agent runs in CI and reads pull request descriptions. An attacker opens a PR. What is the attack, and what stops it?→AI & GPU InfrastructureYour company wants an AI agent in the on-call rotation. What does it do, and what does it never do?→