dev-new-tool

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

• Third-party content exposure detected (high risk: 0.75). The required workflow includes “Search for existing tools” and “Present findings” during Phase 1, which typically involves fetching public web pages at runtime and ingesting their readable text into the agent’s LLM context (public web content / scraped articles as outsider-authored free text).