workflow

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

• Third-party content exposure detected (high risk: 0.80). The skill's examples and required workflow docs (see references/durable-agent-patterns.md's DurableAgent tools like "searchWeb" that call webSearch(query) and SKILL.md examples that use fetch() to external URLs such as https://api.example.com) show the agent invoking web searches/external HTTP endpoints and consuming public web results, which are untrusted third-party content the agent is expected to read and that can directly influence its tool calls and decisions.