virtu_skill

This skill helps you design and optimize Virtu-style execution systems by modeling market microstructure, routing intelligently, and measuring execution

Python

3

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill copyleftdev/sk1llz --skill virtu

SKILL.md19.4 KB

Overview

This skill encodes Virtu Financial–style execution and market microstructure best practices for building high-performance trading systems in Python. It focuses on minimizing market impact, smart venue routing, adaptive execution schedules, and measurable execution quality. Use it to design execution algorithms, smart order routers, and impact-aware trading components.

How this skill works

The skill provides modular components: a market impact model that estimates temporary and permanent impact and builds optimal Almgren–Chriss schedules; a smart order router that scores venues by spread, depth, fill rates, queue advantage, leakage and fees; and execution primitives (VWAP/TWAP slices, passive/ aggressive splitting, IOC routing). Components adapt in real time using live venue states, volume profiles, and randomized sizing to reduce predictability.

When to use it

Building execution algorithms (VWAP, TWAP, adaptive schedules)
Designing a smart order router across multiple venues
Estimating and minimizing market impact before trading large sizes
Implementing anti-gaming and randomization logic to hide intent
Measuring execution quality (slippage, implementation shortfall, fill rates)

Best practices

Model and validate impact on historical and recent intraday data before trading
Prefer passive posting when liquidity allows; use aggressive sweeps only for residuals or urgency
Split large orders across time and venues; avoid single-block execution unless justified
Continuously measure execution quality and feed metrics back into strategy parameters
Incorporate venue-specific rules (fees, rebates, queue behavior) and anti-gaming protections

Example use cases

Implement an Almgren–Chriss based scheduler to produce minute-by-minute execution trajectories
Build a score-based smart order router that allocates slices to venues by spread, depth, and leakage
Create an adaptive VWAP that scales participation with observed volume and limits max participation
Combine passive posting with IOC sweeps to balance price improvement and fill certainty
Simulate and backtest impact-aware strategies to choose risk aversion and urgency parameters

FAQ

Start by backtesting across historical intraday scenarios: higher risk aversion (kappa) front-loads execution and reduces exposure to drift but increases impact. Tune urgency by balancing implementation shortfall versus realized impact on held-out days.

How does the router prevent information leakage?

The router scores venues for leakage and spreads allocations, uses passive posting where possible, randomizes slice sizes/timing, and respects venue depth limits to avoid repeated, predictable patterns that reveal size.