agents-swarm-orchestration

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md "Fact-Checking" section plus data/sources.json explicitly direct the orchestrator to "use web search/web fetch" and list public URLs (e.g., github.com, arxiv.org, openai.github.io), meaning the agent will fetch and read untrusted third‑party web content that can influence its decisions.