prompt-writing

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md instructs the agent to query and inspect user-uploaded/public tunes and the "public gallery" via CLI commands (e.g., "astria tunes list --gallery --model-type faceid --limit 200" and "astria tunes get " to inspect orig_images), which are open/public, user-generated sources the agent must read and use to choose or reference tunes that materially influence prompt construction.