azure-app-onboard-prereq

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md Step 2 (“Scan Workspace”) and Step 3 sub-checks read free text from the user’s repository files (e.g., project manifests, config/source files, Docker Compose, HTML), which the user can freely author/poison, without requiring the agent to first select a specific item.