agentsclimarketplace
Sign in

Regime audit

Skill rgourley/quant-garage/skills/regime-audit

Workflow composite that runs change-point-detector + hurst-exponent on SPY plus the 11 SPDR sector ETFs. Reports per-name the last detected regime shift, current persistence classification (mean_reverting / random_walk / trending), and cross-sector summary (broad_regime_shift / localized_regime_shift / trend_dominated / mean_reversion_dominated / mixed_stable). Requires Stocks Basic. Runs on the free tier.From its SKILL.md

Install
npx -y skills add rgourley/quant-garage --skill regime-audit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.7 KB, 340 tokens by cl100k_base, as published. Nobody here has run it

regime-audit

Runs change-point-detector and hurst-exponent on SPY + 11 SPDR sector ETFs. Reports a matrix view: for each name, when the last regime shift happened, current annualized return + vol per segment, and the Hurst persistence classification.

Answers "where has the market regime shifted, and which sectors are in what regime right now?"

When to invoke

  • Weekly market context review
  • Sector rotation prep
  • "Is this a trending or mean-reverting environment?"
  • The user says "regime audit", "sector regimes", "regime shift map"

What you need

  • MASSIVE_API_KEY exported
  • Stocks Basic minimum

Optional:

  • --tickers (default: SPY + 11 SPDR sector ETFs)
  • --lookback-days (default 504)
  • --lambda-run (default 250) — change-point prior mean run length

What you get back

Layer 1: JSON with per-ticker hurst, hurst_classification, n_change_points, last_change_point_date, last_change_point_confidence, current_segment (annualized return

  • vol), n_segments. Top-level by_regime counts, n_shifted_recently, summary_verdict.

Layer 2: rendered note. Header verdict + summary counts, per-name table, one-line Take.

Foundations used

  • Composes change-point-detector and hurst-exponent
  • Uses massive-api-patterns transitively.

What ships with it: 4 files

2.8 KB alongside SKILL.md

references/

Keep looking

Skills are one crate of 326,871. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.