tamagui

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md explicitly instructs the agent to "Check local first, fall back to GitHub" (https://github.com/tamagui/tamagui), so the skill may fetch and read public GitHub content (user-generated/untrusted) as part of its workflow, which could materially influence agent actions.