agent-friendly-apis

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

• Third-party content exposure detected (high risk: 0.80). The skill's Teaching mode explicitly instructs the agent to fetch and follow lesson content from the public Vercel Academy site (e.g., "GET https://vercel.com/academy/agent-friendly-apis/.md" and "Follow instructions in the block"), meaning the agent will read and act on third-party web-hosted content that can change its next actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's Teaching mode explicitly performs runtime GET requests to https://vercel.com/academy/agent-friendly-apis/.md (and other https://vercel.com/academy/* endpoints) and tells the agent to "Follow instructions in the block", so fetched remote markdown directly controls the agent's prompts and is required for the skill to run.