MCP (Model Context Protocol)
Also known as: Model Context Protocol
A standard way to hand an AI model a set of typed tools it can call. For trading, the server holds the keys and signs orders; the model only sees tools.
MCP lets a model call tools like place_order, cancel, and flatten while the server does the signing, bookkeeping, and safety checks. The model never sees an exchange key, so a leaked transcript never contains one.
Most agent-native entries here expose their surface as MCP, including OKX Agent Trade Kit, Alpaca MCP, and Kraken CLI. The guardrails belong in the server, not the prompt.
Related
- OKX Agent Trade Kit Official open-source MCP toolkit, ~164 tools across spot, perps, options, and grid bots. Keys stay off the model.
- Alpaca MCP Official MCP server for the Alpaca brokerage API. Strong on regulated equities, lighter on crypto-native primitives.
- Kraken CLI Single-binary execution engine, 134 commands, built-in MCP server and paper-trading mode. CEX only.
- Superior Trade Managed execution, backtesting, and a deployment lifecycle behind agent-callable endpoints, with an atomic one-call exit and HIP-3 coverage. Closed core, public OpenAPI and docs.
- Loop engineering Designing the control loop an agent runs in (gather context, decide, act, observe, repeat) rather than hand-writing each prompt. A trading agent is a high-stakes example.
What links here
Alpaca MCP · Bankr / OpenClaw skills · Dwellir hyperliquid-skills · edkdev hyperliquid-mcp · Kraken CLI · Loop engineering · OKX Agent Trade Kit · Senpi · Superior Trade