openai-agents-sdk

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

• Third-party content exposure detected (high risk: 0.90). The skill documentation includes examples that use a WebSearchTool to "Search the web" (references/tools.md) and a deterministic workflow where a search_agent's results are passed to downstream agents for analysis (references/patterns.md), which clearly shows fetching untrusted public web content that the agent reads and that can materially influence subsequent actions.