improve

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

• Third-party content exposure detected (high risk: 0.90). The skill's protocol explicitly spawns research agents to "survey comparable products" and uses "/research" or "/research-fleet" to verify the competitive landscape (SKILL.md Phase 0 and Phase 3), which requires fetching and interpreting public third-party content whose findings drive selection/attacks and documentation changes, exposing the agent to untrusted user-controlled web content.