Agent skill · research science · personamanagmentlayer

trading-expert

Expert-level algorithmic trading, market systems, quantitative analysis, and trading platforms. Use when the user mentions algorithmic trading, quant, markets, finance, or high-frequency trading, or when the task involves Trading Systems, Market Data, or Execution.

Why this skill is useful

Adds executable trading strategy implementations and backtesting scripts that enhance algorithmic trading capabilities.

What it needs

Requires numpy, pandas installed locally. About 7k tokens when loaded. Last updated 2026-03-30. 41 stars on the source repository.

What this skill does

Trading Expert Expert guidance for algorithmic trading systems, quantitative analysis, market data processing, and trading platform development. Core Concepts Trading Systems Algorithmic trading strategies High-frequency trading (HFT) Market making Arbitrage strategies Portfolio optimization Risk management Market Data Order book processing Tick data analysis Market microstructure Real-time data feeds Historical data analysis Execution Order routing Smart order routing (SOR) Execution algorithms (TWAP, VWAP) Slippage minimization Transaction cost analysis Trading Strategy Implementation Order Execution Risk Management Market Data Processing Execution guardrails The order-execution example above stops short of venue connectivity on purpose. sendtovenue is where an illustration becomes a system that can lose money, and a Snyk audit flags this skill as W009 (direct money access) on that basis. Before wiring it to a broker or exchange: Paper trade first. Default to a simulated or paper endpoint and make live routing an explicit, reviewed configuration change. OrderManager(live=True) should never be the default in any environment. Check risk before routing, never after. A routed order cannot be un-sent, and a fill can arrive in microseconds. Pre-trade checks that raise are safer than checks that return a boolean a caller may ignore. Enforce a kill switch. A single operator action must halt all new order submission and cancel resting orders, independently of strategy logic. Bound everything. Maximum order notional, maximum position per symbol, maximum daily loss, and maximum message rate. Breach means stop, not clamp. Use client order IDs idempotently. A retried submission must not create a second order; reject a reused ID rather than routing it again. Never hardcode broker credentials. Load them from a secrets manager, use the narrowest permission the strategy needs, and keep read-only market-data credentials separate from execution credentials. …

How to use it

Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:

@skills personamanagmentlayer/trading-expert

View the source on GitHub

Browse the @skills marketplace