# Macro Regime

> This skill should be used when the user asks to "classify the macro regime", "score expansion versus recession", or create a dated regime signal for capital market assumptions and SAA research.

- **Type:** Skill
- **Install:** `agentstack add skill-nutdnuy-self-driving-portfolio-skill-macro-regime`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [nutdnuy](https://agentstack.voostack.com/s/nutdnuy)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [nutdnuy](https://github.com/nutdnuy)
- **Source:** https://github.com/nutdnuy/self-driving-portfolio-skill/tree/main/skills/macro-regime

## Install

```sh
agentstack add skill-nutdnuy-self-driving-portfolio-skill-macro-regime
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# macro-regime

The 4-regime classifier from Ang/Azimbayev/Kim (2026) §3.2.

## Methodology

Four indicator families are scored on a [-1, +1] scale (positive = "hot",
negative = "cold"). Each family is a z-score of recent readings vs a
rolling 5-year baseline.

| Family                | FRED series                          |
| --------------------- | ------------------------------------ |
| Growth                | INDPRO (industrial production, YoY)  |
| Inflation             | CPIAUCSL (headline CPI YoY)          |
| Monetary policy       | DFF − T10Y3M (real-curve proxy: fed funds minus 10y-3m spread) |
| Financial conditions  | NFCI (Chicago Fed; positive = tight) |

The regime score is a fixed-weight projection (paper §3.2):

```
expansion  =  +growth  −0.5·inflation  −0.5·monetary  −1.0·fci
late_cycle =  +growth  +1.0·inflation  +1.0·monetary  +0.5·fci
recession  =  −growth  −0.5·inflation  +0.0·monetary  +1.0·fci
recovery   =  +growth  −1.0·inflation  −1.0·monetary  −0.5·fci
```

Scores are softmax-normalised across the four labels. The argmax is the
regime label; the full vector is exposed downstream.

## Scripts

- `scripts/fetch_macro.py` — pull an authenticated FRED vintage when
  `FRED_API_KEY` is set; otherwise apply the observation-date cutoff to
  latest-revised public CSV history and disclose that policy in the artifact.
- `scripts/classify_regime.py` — runs the scoring, writes `regime.json`.

## Output contract

`schemas/regime.schema.json`. The `notes` field is filled by the LLM in
agent mode; a templated version is filled when run from CLI.

Run through `pipeline/orchestrator.py` for schema gating and governed output.
Treat direct CLI output as an isolated, ungoverned stage result.

## CLI

```bash
python skills/macro-regime/scripts/classify_regime.py \
       --as-of 2026-05-08 --out outputs/demo01/regime.json
```

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [nutdnuy](https://github.com/nutdnuy)
- **Source:** [nutdnuy/self-driving-portfolio-skill](https://github.com/nutdnuy/self-driving-portfolio-skill)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-nutdnuy-self-driving-portfolio-skill-macro-regime
- Seller: https://agentstack.voostack.com/s/nutdnuy
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.