- Home
- Skills
- Trading & Investing
- Sentinel Quant Risk Desk
Works with the AI Agents you already use
Sentinel Quant Risk Desk
Most AI trading prompts go straight from "this chart looks good" to "buy".
$9
Sentinel Quant Risk Desk
Example session with this skill installed
Scan NVDA, MSFT, AMD, JPM, XOM, LLY, AAPL and CAT for my $25k paper account. Also split $500 a month into long-term ETFs. Which trades pass your risk checks?
- Read your context and instructions
- Compiled the sentinel quant risk
Summary
Scanned 8 names against a $25,000 paper book.
4 approved, 4 blocked. Market is in a NORMAL regime, so full position sizing applies, but Technology is one of the three weakest sectors and its names carry a confidence penalty.
Market regime
| Metric | Reading |
|---|---|
| Regime | NORMAL (VIX 19.3) |
| Size multiplier | 1.0x |
| SPY today / 20 days | -0.36% / -1.16% |
| Strong sectors | Financial Services, Healthcare, Consumer Cyclical |
| Weak sectors | Technology, Energy, Communication Services |
VIX is under 22, so short-term momentum buys are allowed. A move above 22 would cut every new position to 60% size.
Approved paper trades
| Ticker | Sleeve | Shares | Notional | Entry | Stop | Target | Confidence |
|---|---|---|---|---|---|---|---|
| LLY | mid (trend) | 1 | $1,444 | 1443.50 | 1290.63 | 1749.23 | 0.93 |
| AAPL | mid (trend) | 7 | $2,494 | 356.27 | 327.95 | 412.92 | 0.82 |
| JPM | mid (trend) | 7 | $2,423 | 346.19 | 322.82 | 392.93 | 0.64 |
| AMD | short (momentum) | 2 | $1,068 | 534.06 | 438.25 | 677.78 | 0.68 |
LLY: textbook trend. Price above its 20, 50 and 200-day averages, 5 of 5 entry confirmations, and Healthcare is a top-3 sector (+0.05 confidence).
AAPL: strong trend score (86.7) but in a weak sector, so the gate trimmed 0.05 off confidence. Still clears 0.60 comfortably.
- JPM: steady trend in the strongest sector. Lowest conviction of the four, sized by its ATR stop.
AMD: momentum setup with only 3 of 5 confirmations. Approved, but watch the 438 stop closely; it is the widest relative risk on the sheet.
Blocked ideas
| Ticker | Failed checks | Why |
|---|---|---|
| NVDA | G08, G12 | Earnings in 3 days (2026-11-17), inside the 5-day short-sleeve blackout. Momentum and trend scores both 0. |
| MSFT | G12 | Confidence 0.56 after the weak-sector penalty, below the 0.60 floor. Weak short-term momentum (10.5). |
| XOM | G12 | Confidence 0.42. Energy is a bottom-3 sector and the trend is middling. |
| CAT | G12 | Only 1 of 5 entry confirmations. No setup right now. |
Long-term DCA ($500/month)
| ETF | Weight | Amount |
|---|---|---|
| VTI | 40% | $200 |
| VXUS | 20% | $100 |
| QQQ | 15% | $75 |
| SCHD | 15% | $75 |
| BND | 10% | $50 |
What would change the call
- VIX above 22: all new sizes cut to 0.6x. Above 30, short-term buys stop entirely.
- NVDA after 2026-11-18: re-scan once the earnings blackout clears.
- MSFT: becomes eligible if Technology climbs out of the bottom three sectors or momentum improves.
- Total new exposure if all four fill: $7,429 (about 30% of the book), well inside the 90% gross cap and 30% sector cap.
Next step: run review daily on your portfolio file to catch stop, target and time exits.
Paper analysis only. Not financial advice.
Connects securely to your tools. The creator never sees your data.
About this skill
Most AI trading prompts go straight from "this chart looks good" to "buy". Sentinel Quant Risk Desk puts a deterministic risk engine between your agent's opinion and any trade. Your agent writes the thesis and its confidence; a Python governance gate decides whether the idea survives and how big it may be.
It is the governance layer of Sentinel Quant, a multi-strategy AI trading system described in a published paper (DOI 10.5281/zenodo.19284847), packaged so any SKILL.md agent can run it.
What it does
- Market regime: classifies LOW_VOL / NORMAL / HIGH_VOL / CRISIS from VIX and scales position size to match.
- Sector rotation: ranks the 11 SPDR sector ETFs by 20-day return and nudges confidence toward strong sectors and away from weak ones.
- Two trading sleeves: short-term momentum (days) and mid-term trend (weeks), each with its own position limits, ATR stops, targets and max holding period.
- 18 governance checks: kill switch, daily trade limit, duplicate orders, daily loss cap, sleeve limits, regime gate, SPY circuit breaker, earnings blackout windows, 60-day correlation block, entry-quality confirmations, confidence threshold, loss-streak sizing, cash reserve, gross exposure cap, sector concentration and minimum trade size.
- Risk-based sizing: 1% of equity at risk to an ATR stop, capped per sleeve, cut in volatile regimes and after losing streaks.
- Position review: flags stop, target, time and earnings exits on a paper portfolio and restates the original thesis so the agent judges whether it broke.
- Long-term DCA: optional monthly split across VTI, VXUS, QQQ, SCHD and BND.
Why buyers like it
- Blocked ideas come with reasons. You see exactly which rule stopped a trade ("earnings in 3 days", "0.91 correlation with a holding", "Technology at 34% of book").
- No API keys, no accounts. Uses free Yahoo Finance data through yfinance.
- Transparent and editable. Every threshold sits in one
POLICYdict; ask your agent to tighten or loosen any rule. - Safe by design. Analysis and paper trades only. It never connects to a broker or places orders.
Requirements
Python 3.9+ and pip install yfinance pandas numpy.
Educational and research tool. Not financial advice.
Paper analysis only. No live execution or broker APIs. Requires local Python environment with yfinance. Uses example risk parameters that may need manual tuning in the script.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Requires