daishin-report-search

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and parses public HTML reports from a third‑party GitHub Pages mirror (via the GitHub recursive tree API at https://api.github.com/repos/jay-jo-0/github_pages_repo/... and raw content at https://raw.githubusercontent.com/... / the GitHub Pages URL), ingesting untrusted user-hosted page content which the agent reads and uses to produce JSON/summaries, allowing malicious page content to influence outputs.