brutal-self-review

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (Architectural Review) instructs the agent to "load its SKILL.md" and "follow its references" for the how-to-golang skill, and the references/go-ecosystem.md explicitly points to an external public URL pattern (https://context7.com/…/llms.txt), meaning the agent is expected to fetch and act on open/public third‑party content that can influence decisions.