hedging

Upsert one daily hedging decision into results/hedging using offline MCP data.

Updated May 3, 2026
One-click install
npx skills add https://github.com/CYP0630/agentic_benchmark_single_agent --skill hedging-cyp0630
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hedging
Source: https://github.com/CYP0630/agentic_benchmark_single_agent/tree/main/skills/hedging
Command: npx skills add https://github.com/CYP0630/agentic_benchmark_single_agent --skill hedging-cyp0630

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires duckdb, numpy, pandas, fastmcp, pydantic, and includes scripts (resource) components.

What problem does it solve?

Makes one daily hedging decision for a fixed ordered stock pair, or selects the fixed pair on the run's first day, by querying offline data through the hedging_mcp server. Each invocation handles exactly one target date and upserts one recommendation into results/hedging/. The same skill can be driven by an external date-loop for any contiguous run window (benchmark, live, or otherwise).

Use this skill whenever the user asks for a hedging decision, pair selection, or one daily hedging step.

Core Features & Use Cases

  • Offline MCP data access for deterministic pair selection and daily hedging for a fixed pair.
  • Upserts a single daily hedging decision into results/hedging/ with fields for dates, prices, actions, and status.
  • Supports both initial pair selection on day one (IS_FIRST_DAY) and fixed-pair execution when provided.

Quick Start

Instruct the hedging system to pick today's hedge for the configured pair using offline MCP data and upsert the result into the daily run file.

Frequently Asked Questions about hedging

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate daily pair trading hedging decisions using offline data?

Daily pair trading hedging decisions are automated by querying offline MCP data for a fixed stock pair and upserting one deterministic recommendation into a structured JSON file. The system processes exactly one target date per invocation to enforce no-look-ahead bias.

What is deterministic hedging sizing and how does it prevent look-ahead bias?

Deterministic hedging sizing calculates position sizes using only data available up to the target date. The system enforces no-look-ahead bias by processing one day at a time through scripts like upsert_hedging_decision.py, ensuring reliable single-day decisions.

Can I use DuckDB and Pandas for offline MCP pair selection on the first day?

Yes, offline MCP pair selection on the first day utilizes DuckDB and Pandas to query historical prices, news, and filings. The skill selects a fixed ordered stock pair on day one (IS_FIRST_DAY) and executes fixed-pair hedging for subsequent days.

How do I run a contiguous multi-day hedging benchmark with a single-day decision script?

You run a contiguous multi-day hedging benchmark by driving the skill with an external date-loop. Each invocation handles one target date and upserts its recommendation into results/hedging/, allowing the same skill to process any contiguous run window for backtesting or live trading.

Does the hedging skill require FastMCP and Pydantic to fetch prices and filings?

Yes, the skill requires FastMCP and Pydantic to interact with the hedging_mcp server and fetch offline prices, news, and filings. These dependencies ensure structured data validation and reliable tool interaction for daily hedging decisions.

What are the limitations of using a fixed stock pair for daily hedging decisions?

The limitation of fixed stock pair hedging is that pair selection only occurs on the run's first day. Subsequent days execute decisions for that fixed pair without re-evaluating alternatives, which simplifies automation but restricts dynamic adjustment to changing market conditions.