alpaca-broker-trading-orders

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to fetch live documentation (e.g. https://docs.alpaca.markets/llms.txt and https://docs.alpaca.markets/llms-full.txt, or an alpaca-docs MCP) to obtain endpoint schemas and guidance that would be injected into prompts and used to generate code that can place or modify trades, so these URLs are runtime dependencies that directly control agent behavior.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly a broker trading API: it provides endpoints to create, modify, replace, and cancel orders (POST /v1/trading/accounts/{id}/orders, PATCH, DELETE), place market/limit/notional (fractional) orders, close positions, and read trading-account fields like buying_power. It is explicitly designed to execute market orders and manage positions on behalf of user accounts (account_id in path), so it grants direct financial execution capability.