cli-anything-web-yu-pri

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime path uses Playwright to open Japan Post pages (e.g., open-login, status, plan, contents fill, and snapshot), and the agent may ingest browser-extracted HTML/text from the outsider-authored public web content at https://mgr.post.japanpost.jp/... into the LLM context.