review-skills

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

• Third-party content exposure detected (high risk: 0.70). The skill instructs the agent to download and read SKILL.md files from the public GitHub repo at https://github.com/anthropics/skills via scripts/download_anthropics_skills.py (cached under ./.cache/anthropics-skills), so the agent will fetch and interpret open/public third-party content as part of its workflow.