prompt-improver

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

• Third-party content exposure detected (high risk: 1.00). The skill's core Research phase (SKILL.md Phase 1 Research and references/research-strategies.md "Web Research for Best Practices"/"WebFetch" and "WebSearch" steps) explicitly instructs the agent to fetch and read online documentation and web search results, meaning it will ingest untrusted public web content that can influence questions and execution.