token-saving-techniques

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly directs the agent to fetch additional skills from public third-party repositories (https://skills.sh/viktor-ferenczi/skills and https://github.com/viktor-ferenczi/skills), meaning the agent would ingest and act on untrusted, user-hosted content that could influence its actions.