What Quote.Trade provides for AI agents
Quote.Trade publishes a hosted MCP endpoint, OpenAPI, REST and WebSocket documentation, agent guides, and open-source clients. These interfaces let agents discover tools and use structured data; they do not promise profitable strategies.
Public read-only agent tools
An agent can retrieve platform status, exchange information, instruments, ticker data, market depth, and an API manifest. These tools provide a low-risk way to test connectivity and build research or monitoring workflows before an account is funded.
Account and order tools with server-side controls
Authenticated tools can read approved account data, preview orders, and submit trades. Keep API credentials on the server and enforce account, symbol, order-size, leverage, and approval limits outside the model.
MCP, REST, and WebSocket roles
Use MCP when an agent needs discoverable tools and an approval flow. Use REST for direct signed API calls. Use WebSocket for live market, order, account, position, and risk updates. A production agent may use all three.
Keep credentials and limits outside the model
Let the model propose a structured order. Validate it in code, apply account and risk limits, show the exact order for approval, submit it once, and confirm the result from Quote.Trade. The model never needs the API secret.