desk-operating-model

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

• Third-party content exposure detected (low risk: 0.10). The skill describes an AI agent team reading market data, news, and research files, which represents low-risk ingestion of external web content and internal documentation without direct outsider submission paths like email or chat feeds.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (medium risk: 0.30). The skill documents a trading desk operating model ("HyperGrok") involving live financial trading ("Mainnet desk") on Hyperliquid via an API wallet (SKILL.md:24, SKILL.md:46, SKILL.md:54). This constitutes a direct financial operation (trading/execution). Access control is enforced via Grok Bot's secure secret store for API keys and its built-in execution/approval controls (SKILL.md:46, SKILL.md:70, SKILL.md:72), meeting the independent authentication/authorization criteria.