Guide · Agent skill
How to use the Hedging Strategy Design agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The hedging-strategy skill is a set of instructions and reference code that lets an AI agent help you plan how to protect an existing investment portfolio from losses. It is written mainly around China A-share instruments such as CSI 300 futures, 50ETF options and the gold ETF 518880.SH. After this guide you will know what the skill does, how to install it, how to phrase a request that triggers it, and what to check before acting on its numbers.
Hedging Strategy DesignHKUDS/Vibe-Trading · MIT Download the skill Quick start
- 1Download the ZIP. Download the hedging-strategy ZIP from this page.
- 2Unzip into a skills folder. Unzip it into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder.
- 3Or upload in the Claude apps. Go to Settings → Capabilities → Skills and upload the ZIP.
- 4Install Python packages. Make sure numpy, scipy and pandas are available for the example calculations.
- 5Describe your position. State the portfolio size, market, beta and what risk you want to protect against.
- 6Request a comparison. Ask for hedge ratios, cost estimates and an execution plan across the options the skill covers.
- 7Verify before acting. Check current prices, margin and contract specs, and have a qualified professional review the plan.
What the hedging-strategy skill makes the agent do
An agent skill is a folder containing a SKILL.md file that tells the agent how to handle a certain kind of task. When this skill is active, the agent follows a hedging design process for a position you already hold. Its core idea is that hedging does not eliminate risk; it exchanges unknown losses for known costs.
The skill covers linear hedges (futures or ETFs that offset market risk), nonlinear hedges (options such as protective puts, collars and put spreads), tail-risk protection against rare crashes, and cross-asset mixes of stocks, bonds and gold. For each approach it describes how to calculate a hedge ratio, estimate cost and lay out an execution plan.
- Beta hedging with index futures or ETFs, while keeping stock-specific exposure
- Protective puts, collars and put spreads, with parameter tables
- Far out-of-the-money put tail-risk hedges
- Hedge-ratio methods: OLS, minimum variance and EWMA
- Cost comparison across futures, options and cross-asset hedges
Install the hedging-strategy skill
You can install the skill in one of three places. For a personal setup, download the ZIP from this page and unzip it into ~/.claude/skills/. To limit it to a single project, unzip it into that project's .claude/skills/ folder. In the Claude apps, you can instead upload the ZIP under Settings → Capabilities → Skills.
After unzipping, the skill should be a folder that contains SKILL.md. The source material for this skill ships as that single file.
- Personal install: ~/.claude/skills/
- Project install: the .claude/skills/ folder inside your project
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
~/.claude/skills/Prepare Python and your data before running calculations
The skill runs scripts, and its worked examples use Python with numpy, scipy and pandas. The file does not state these requirements explicitly, so you should confirm they are installed before asking for hedge-ratio calculations. The skill does not connect to the internet, so you must supply the price and return data yourself. No data source is specified in the file.
Your data needs the daily or periodic returns of your portfolio and of the hedge instrument you are considering. The beta and hedge-ratio formulas in the skill work on those return series.
- Python with numpy, scipy and pandas installed
- Return series for your portfolio, provided by you
- Return series for the hedge instrument (for example CSI 300 futures or a 50ETF put)
Trigger the skill with natural requests
You do not need special syntax. Describe your position and what you want to protect against, and mention hedging, protection, tail risk or beta. The agent should then use the skill's framework rather than generic advice. Give the size of the holding, the market it trades in and your tolerance for cost or upside give-up, because the skill's rules depend on these.
Good requests name the portfolio, the risk and the constraint. Vague requests produce vague plans.
- "I hold a 10 million RMB A-share portfolio with beta about 1.2. Compare CSI 300 futures and a CSI 300 ETF hedge and estimate annual cost."
- "Design a collar on 1 million shares of 50ETF. I can give up upside above 5% but want near-zero cost."
- "Calculate the minimum-variance hedge ratio from these two return series and tell me whether an OLS or EWMA estimate fits a weekly rebalance."
- "My portfolio is mostly stocks. Compare a 60/40 stock-bond mix with adding gold ETF 518880.SH for tail risk."
Read the output and check the assumptions
The skill's output is a set of hedge ratios, cost estimates and an execution plan. Many of its figures are illustrative rules of thumb, not current market data. The file gives example option premiums of 1-3% per month and margin of about 12% for CSI 300 futures. These may be outdated, and the skill does not check current prices or contract specifications.
Before acting, confirm the contract multiplier, margin, strike prices and premiums with your broker or the exchange. Treat the agent's plan as a draft to review.
- Verify prices, margin and contract specs from current sources
- Check that the example rules of thumb still hold
- Have a qualified professional review any real hedge
Safety notes on scripts and network use
The skill runs scripts but does not use the network, according to its listing. Scripts run on your machine and work only on the data you provide. Review any script before it runs, particularly if you adapt it.
Derivatives can lose more than the hedge protects. Hedging adds cost and can fail when correlations change; the source notes that stock-bond correlation is not stable and that the traditional 60/40 mix failed in 2022 in the US.
- Scripts run locally; the skill does not use the network
- Hedges can cost more than they save and can fail in unusual markets
- The skill is a planning aid, not a licensed investment recommendation
What people use it for
- Estimating how many CSI 300 futures contracts would offset the market risk of a stock portfolio.
- Comparing a protective put, a collar and a put spread on an existing ETF position.
- Drafting a hedging proposal for a risk manager to review.
- Learning beta hedging and minimum-variance hedge ratios with worked examples.
- Comparing a stock-bond mix with a gold ETF allocation for tail-risk protection.
Frequently asked questions
How do I install the hedging-strategy agent skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. In the Claude apps you can instead upload the ZIP under Settings → Capabilities → Skills.
What do I need to run the hedging-strategy skill?
The skill's example calculations use Python with numpy, scipy and pandas, though the file does not state this explicitly. You also need to supply your own price and return data, since no data source is specified.
Is the hedging-strategy skill free and what is its licence?
The source is listed under the MIT licence. The skill files are part of the HKUDS/Vibe-Trading repository, and you should read the licence there for the full terms.
Does the skill use the internet to get live prices?
No. The skill is listed as not using the network, so it does not fetch current prices. You provide the data, and the skill does not check current market prices or contract specifications.
Can I use this skill for US stocks?
Most worked examples, instruments and codes are specific to China A-shares, so US or other markets need separate verification. The skill's VIX discussion is marked as a US equities example, but the rest should be checked against your own market.
How does this compare with asking a general chatbot about hedging?
The skill gives the agent a structured process: identify the risk, choose the instrument, calculate the ratio, then evaluate cost. It includes specific formulas and parameter tables, but its numbers are illustrative and must be verified against current data.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
This skill suits investors or analysts with a stock portfolio who want a structured comparison of hedging options, and students learning beta hedging or option collars. It is less suitable for live trading without review, for personal investment advice, or for markets outside China A-shares without independent checking of the figures.
Download the skill