send-to-minion

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires keeping the user's original prompt intact and including it verbatim in the minion task packet, which would force the LLM to read and forward any secrets (API keys, tokens, passwords) present in the prompt.

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

• Third-party content exposure detected (medium risk: 0.30). This skill is invoked only when the user explicitly requests delegation ($send-to-minion), and at runtime it passes the user’s prompt directly into a spawned subagent via the delegation tools, so outsider-authored free text is ingested before any item selection.