cta-firehose — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited cta-firehose (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Daily radar for systematic-flow trigger levels — a public-data replication of the sell-side CTA models (GS Futures Strats / Nomura McElligott style) that drive headlines like "SPX 7230 = first CTA sell trigger".
| Signal | Mechanism | Trade meaning |
|---|---|---|
| Momentum flip lines (1M/3M/6M/12M) | Price vs N-trading-days-ago close | Cross below = that horizon's trend signal flips negative = mechanical selling tranche |
| 50DMA (±1%) | Bank models' first-trigger zone clusters here | The "7230" type number |
| 20/100/200 DMA | Secondary tranches / long-term line | 200DMA break = trend funds flip short |
| 1M/3M realized vol | Vol-control & risk-parity sizing | RV > 20% = mechanical de-risking regardless of direction |
Indices: SPX (^GSPC), NDX (^NDX), RTY (^RUT). Edit INDICES in scripts/cta_levels.py to add more (e.g. ES=F, ^SOX).
CTA AUM (~$300-400B) runs near-identical multi-horizon momentum models, so trigger levels cluster industry-wide. A simple replication lands within ~0.5% of the bank numbers (validated 2026-06-11: our 50DMA 7195 vs GS first trigger ~7230). Vol-targeting then amplifies: selling → RV up → forced de-leverage → more selling.
Daily post-close, weekdays (GitHub Actions, .github/workflows/cta-firehose.yml):
21:15 UTC (17:15 ET) — settle prices in, levels for tomorrowState (scripts/state.json) remembers which side of each trigger every index closed on; a cross since the last run prepends a 🔔 CTA TRIGGER BREACHED banner so the Telegram alert is impossible to miss.
# Print only
uv run --with yfinance --with numpy python scripts/cta_levels.py
# Dry-run the Telegram send
TEST_MODE=1 uv run --with yfinance --with numpy python scripts/cta_levels.py --telegram🚨 1M动量线 7399 (-0.2%) ← short-horizon CTA already selling
✅ 50DMA 7195 (+2.7%) ⭐首轮触发区 ← THE level; below = first big tranche
✅ 3M动量线 6781 (+8.9%) ← second tranche
✅ 200DMA 6868 (+7.5%) ← trend funds flip short below
RV: 1M 12.6% / 3M 14.9% 🆗 ← >20% = vol-control selling on topCross-check against the leaked bank numbers (@Michael_QQQ2025, The Market Ear, ZeroHedge reposts of McElligott) — our lines are the skeleton, theirs add positioning size ($bn per tranche).
close date it used.
(how much is left to sell) — pair with the Goldman/Nomura leaks for size.
run keeps them honest.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.