Guide · Agent skill
How to use the Sector Rotation Analysis agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The sector-rotation skill is a methodology guide that teaches an AI agent how to run a sector rotation review of China A-share industries using the Shenwan (Shenwan Hongyuan) industry classification. Agent skills are folders of instructions that an agent such as Claude loads when a request matches them. This guide shows you how to install the skill, how to ask for an analysis, and what the output looks like, along with its limits. After reading it you will be able to get a structured sector review and judge whether it fits your work.
Sector Rotation AnalysisHKUDS/Vibe-Trading · MIT Download the skill Quick start
- 1Download the ZIP. Download the sector-rotation ZIP from this page. It contains SKILL.md.
- 2Unzip into a skills folder. Place the sector-rotation folder in ~/.claude/skills/ or in .claude/skills/ inside your project.
- 3Or upload in the Claude apps. Open Settings, go to Capabilities, then Skills, and upload the package.
- 4Supply your data. Provide A-share industry returns, earnings and fund-flow data, since the skill does not fetch them.
- 5Ask for a review. Describe the sectors, lookback window and output you want in plain language.
- 6Verify the output. Check the scores, rankings and weights against your own data before acting on them.
What the skill makes the agent do
Once installed, the agent can produce a structured sector-rotation review of A-share industries. It scores each sector out of 100 across five dimensions: earnings growth (30%), earnings trend (20%), business-cycle indicators (20%), policy support (15%) and valuation safety (15%). It also ranks sectors by price momentum and earnings momentum, and suggests which sectors to overweight or underweight.
The skill also covers supply-chain transmission patterns (for example, from lithium carbonate to batteries to electric vehicles), a comparison framework for valuation, profitability and fund flows, and a report template. The file is written mostly in Chinese and gives rules and a sample formula rather than a live data feed.
Install the sector-rotation skill
Download the ZIP file from this page. The skill package contains one file, SKILL.md. Unzip it so that the folder sits in one of these locations: ~/.claude/skills/ for your personal skills, or .claude/skills/ inside a project folder for skills that only apply to that project.
You can also upload the skill in the Claude apps. Go to Settings, then Capabilities, then Skills, and upload the package there.
- Personal: ~/.claude/skills/sector-rotation/
- Project: .claude/skills/sector-rotation/ inside your project folder
- Claude apps: Settings → Capabilities → Skills → upload
~/.claude/skills/sector-rotation/SKILL.mdTrigger the skill with a natural request
You do not need a special command. Describe the task in plain language that mentions China A-share industries, sector rotation, industry momentum or overweight and underweight decisions. The agent matches your request against the skill's description and loads its rules.
Be specific about the sectors, time window and data you have. The skill expects you to supply or point to the industry return, earnings and fund-flow data, because the file does not include a data provider.
- Rank the Shenwan first-level industries by momentum over the last 60 trading days, skipping the most recent 5.
- Score the electronics, automotive and communications sectors on the five-factor framework and suggest overweight or underweight ranges.
- Explain how a lithium carbonate price rise is likely to pass through the battery and EV supply chain.
- Give me a sector rotation review using the report template, with ETF examples for each recommended sector.
What you need before you start
The momentum example is written in Python and uses pandas (the code calls pd.DataFrame and pd.Series), so Python with pandas is needed if you want to run it. The skill does not include code to download market data.
You must supply A-share industry return, earnings and fund-flow data yourself, or connect the agent to another tool that provides it. The file names example sources such as northbound flows (沪深港通), margin balances (融资融券), PMI and ETF subscriptions, but it does not say which provider to use.
- Python with pandas for the momentum example
- A-share industry return, earnings and fund-flow data supplied by you or another tool
Read the output and check it
The skill defines a report layout: a ranking table of the top 10 sectors with a score, change and core logic; a configuration table with overweight, neutral and underweight groups and suggested weight ranges; supply-chain opportunities; and risk notes. Treat the weights and sector views as a starting point for your own research.
The file itself warns about several traps. Low PE can mark a cyclical peak, so price-to-book is safer for cyclicals. Policy shifts can outweigh fundamentals. Themes such as AI concepts can distort momentum signals. Earnings data lags by one to two months, while high-frequency indicators such as PMI are more timely.
- Cyclical stocks: low PE can be a peak signal; use PB as well.
- Policy turns can matter more than fundamentals in A-shares.
- Short-term themes can distort momentum rankings.
- Earnings reports lag by one to two months.
- Rebalance monthly or quarterly rather than often.
What people use it for
- A student learning how momentum and industry-cycle analysis are structured, using the scoring rules as a worked example.
- An analyst drafting a monthly China A-share sector review in the template's table format.
- A developer building a Chinese equity research workflow who wants a starting template for scoring and momentum ranking.
- An investor comparing the attractiveness of Shenwan sectors on valuation, earnings momentum and fund flows.
Frequently asked questions
Is the sector-rotation skill free?
The skill is published under the MIT licence, as stated on this page. Check the licence file in the source repository for the full terms before you reuse or redistribute it.
Does the skill need internet access or run scripts?
No. According to this listing, the skill does not run scripts and does not use the network. It is a set of written instructions and a sample Python function that you run yourself.
Where do I install an agent skill?
Put the sector-rotation folder in ~/.claude/skills/ for personal use or in .claude/skills/ within a project. You can also upload it in the Claude apps under Settings, then Capabilities, then Skills.
Does it work for US stocks or other markets?
No. The classification, ETF codes and data sources are China-specific, built around Shenwan industries and A-shares. It is not designed for other markets.
Why does the agent not return live sector data?
The skill is a methodology guide, not a data feed. It names sources such as northbound flows and PMI but includes no code to fetch them, so you must provide the data or connect another tool.
Can I use this skill for actual trading decisions?
The file gives scoring rules and example sector views, not validated backtests. Verify every conclusion with independent data and your own judgement before making trades.
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
Use this skill if you want a structured, repeatable framework for comparing China A-share industries, or a starting template for a Chinese equity research workflow. Pick something else if you need live market data, automatic retrieval, non-Chinese markets, or backtested trading signals, since the skill provides none of these.
Download the skill