ml-ops

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

• Third-party content exposure detected (high risk: 0.70). The skill's "Companion check" in SKILL.md explicitly instructs the agent to run shell commands and to offer installing missing recommended skills via "npx skills add ..." (which fetches packages from public registries/GitHub), meaning the agent may fetch and ingest third-party code that can influence its tools and behavior.