How to use the Macroeconomic Analysis agent skill (macro-analysis)

What is an agent skill and what does macro-analysis make the agent do?

An agent skill is a folder containing a SKILL.md file. The file tells an AI agent how to handle a certain kind of task. The macro-analysis skill's description is: macroeconomic cycle positioning and central-bank policy interpretation, including GDP, CPI, PMI, rates and FX analysis, with output in the form of major-asset allocation tilts.

Once installed, the agent follows a four-step framework from the file: collect core indicators from the latest three months, decide where the economy sits in the cycle, assess policy impact, and then propose an asset allocation tilt. Its outputs follow a fixed report template with a data snapshot, cycle assessment, central-bank analysis, allocation table and risk warnings.

• Growth indicators: GDP YoY, manufacturing PMI, industrial production, retail sales, fixed asset investment
• Inflation indicators: CPI YoY, PPI YoY, core CPI, M2 YoY
• Rates and FX: 1Y/5Y LPR, DR007, 10-year government bond yield, USD/CNY, US 10Y-2Y spread
• Cycle model: recovery, overheat, stagflation, recession, based on a Merrill Lynch clock framework
• Central-bank frameworks: FOMC statement, dot plot and minutes for the Fed; PBOC toolbox; ECB HICP and PMI

How do you install the macro-analysis skill?

Download the ZIP for macro-analysis from this page and unzip it. The skill is a folder that contains SKILL.md. You can install it in one of two places, depending on how you want to use it.

For a personal install that applies everywhere, unzip the folder into ~/.claude/skills/. For a project-only install, unzip it into the .claude/skills/ folder inside that project. In the Claude apps, you can instead upload the skill under Settings → Capabilities → Skills.

• Personal: ~/.claude/skills/macro-analysis/
• Project: .claude/skills/macro-analysis/ inside your project folder
• Claude apps: Settings → Capabilities → Skills, then upload

How do you trigger the skill with natural requests?

You do not need a special command. Describe what you want in plain language that touches on macro data, central banks or cycle positioning. The agent should recognise the request from the skill's description and apply its framework.

Be specific about the economies and the question. The skill covers China, the United States and Europe, and its output is most useful when you ask for a snapshot, a cycle stage, a policy read or an allocation view.

• "Use the macro analysis skill to place the China economy in a cycle stage using the latest PMI, CPI and PPI."
• "Interpret the latest FOMC statement and tell me whether the Fed sounds hawkish or dovish."
• "Give me a macro briefing on the US, China and Europe with a data snapshot and asset allocation tilts."
• "Read this PBOC statement and tell me what policy tools it implies."

What output should you expect?

The skill asks for a report with a fixed structure. It starts with a snapshot table of core indicators showing latest value, previous value and trend, followed by the cycle positioning, central-bank analysis, an allocation tilt table and risk warnings. Because the template is fixed, you can compare briefings from different months side by step.

The thresholds in the file are generic rules of thumb. For example, it treats China GDP above 5% as normal and below 4% as weak, and it treats a 10-year government bond yield below 2.5% as loose. Treat these as starting points for your own checks, not as validated signals.

What does the skill need to run?

The skill lists no requirements. It does not run scripts and it does not use the network. It provides no live data feed: the file only tells the agent which indicators to collect and where they are normally published, such as the National Bureau of Statistics, Caixin and the PBOC. You must supply the latest figures yourself, either by pasting them into the chat or by providing them in files the agent can read.

• No listed requirements
• Does not run scripts
• Does not use the network
• You provide the current data; the skill does not fetch it

What are the safety limits of this skill?

The skill outputs asset-allocation tilts such as overweight or underweight recommendations for China A-shares, US equities, bonds, commodities and cash. These are directional views, not trade instructions. The file itself notes that macro analysis gives directional judgment rather than exact levels or timing.

Do not rely on it for personal financial decisions without professional review. The file gives no data sources or backtests for its thresholds, so verify any conclusion independently before acting on it.

• Thresholds are generic rules of thumb, not validated forecasts
• The file gives no backtests
• Not suitable as personal financial advice
• Check any figure against the official release before using it

What tips help you get better results?

Give the agent the latest three months of data for each indicator you care about, since the framework relies on trends rather than single readings. Ask for the marginal change, because the file stresses that direction and speed of change matter more than absolute levels.

Remember that data is released with lags. PMI is the timeliest, arriving at the start of the month, while GDP is the most delayed, arriving around 15 days after quarter-end.

• Supply three months of data per indicator
• Ask about direction and speed of change
• Check release timing: PMI early in the month, GDP about 15 days after quarter-end
• Compare reports over time using the same template