using-claude-code-as-pm

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to perform web searches and ingest public third‑party content for competitor and pricing research (see references/RESEARCH-SYNTHESIS.md and references/PROMPT-TEMPLATES.md where "[競合名] をWebサーチで調査" and similar prompts are required), meaning untrusted public webpages are read and used to drive analysis and product decisions.