add-guard-protection

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs scanning and guarding "tool results" and web/search tool inputs (see SKILL.md Step 4 "Scanning tool results for injection" and the JavaScript/Python references showing detectPromptInjection/piRule(query) and the tools.search_web example), which means the agent is expected to ingest untrusted, public tool/web output and let that content influence guard decisions and subsequent actions.