agent-package-manager

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflows (SKILL.md Step 3 and the referenced files like references/manifest-and-lockfile.md and command-workflows.md) explicitly instruct using apm install / apm mcp commands to fetch packages and individual agent/skill files from public GitHub and other external registries (e.g., owner/repo or github/.../agents/*.agent.md), which will deploy prompts/agents/skills that the agent will read and that can change its behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The troubleshooting section includes runtime installation commands that fetch and execute remote scripts (curl -sSL https://aka.ms/apm-unix | sh and irm https://aka.ms/apm-windows | iex), which constitute remote code execution used to install a required dependency (apm).