gr-competitor

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly fetches and extracts content from arbitrary public websites (actionbook extract , blog/index pages, landing/pricing pages and Wayback Machine) and then ingests and synthesizes that untrusted third‑party content into reports and recommendations (see the SKILL.md calling actionbook extract and the prompt templates), so external page content could indirectly inject instructions affecting agent behavior.