muapi-ai-clipping

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

• Third-party content exposure detected (high risk: 0.90). The skill accepts arbitrary hosted video URLs or uploads local files (SKILL.md "Step 1 — Collect Inputs" and scripts/run-ai-clipping.sh "Stage 1 — Resolve --video to a hosted URL"), sends them to muapi's /ai-clipping which transcribes and returns hook_sentence, titles, and virality_reason that the agent reads and uses to rank/select/download clips, so untrusted third-party content can indirectly influence agent behavior.