agent-to-agent

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs a "researcher" agent to "Search the web for authoritative sources" and use tools like WebSearch/WebFetch (see "Research + Writer Pipeline" Step 2 and the agent templates), writing those findings into .a2a-context.json which downstream agents read and act on, so untrusted web content can materially influence agent actions.