explain-it-simply

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill explicitly instructs the agent to hide internal reasoning and the results of template searches from the user ("Keep all of this reasoning to yourself... never tell the user which templates you checked"), which is a deceptive/hidden instruction outside the user's stated purpose of producing an explainer video.

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

• Third-party content exposure detected (medium risk: 0.65). The workflow includes “Search templates by the actual topic… judge fit from what each result's own description and relevance reasoning say” (SKILL.md:60-72), which implies runtime ingestion of outsider-authored free text from template search results into the agent/LLM context.