Fund Manager
Portfolio VaR Report for the Board — Generated in 2 Minutes
Key Takeaway
My AI agent generates a complete risk report — VaR, CVaR, Sharpe ratios, and historical stress tests across equities, crypto, DeFi, and prediction markets — formatted for board presentation in under 2 minutes.
The Problem
Every quarter, the board asks the same question: "What's our risk exposure?"
And every quarter, the answer involves two days in Excel. Pulling prices. Calculating returns. Running Monte Carlo. Formatting tables. Making it look presentable. Then someone asks "what about crypto?" and you start over because it wasn't in the model.
PyratzLabs isn't a traditional portfolio. We have US equities (TSLA, NVDA, PLTR, GOOGL), Japanese equities, French equities, crypto holdings, DeFi positions, and prediction market bets. Try getting Excel to correlate all of that into a coherent risk number.
I needed one command. One report. Every asset class. Board-ready.
The Solution
Warren, my CFO agent, uses the Risk Metrics Calculation skill to generate institutional-grade risk reports across all sub-portfolios. Daily VaR, weekly VaR, CVaR (expected shortfall), Sharpe ratios, Sortino ratios, maximum drawdown analysis, and historical stress scenarios — all in one pass.
The output is a formatted markdown report that drops straight into the board pack. No Excel. No manual formatting. Two minutes.
The Process
The config defines the portfolio structure and risk parameters:
yamlShow code
# portfolio-risk-config.yaml
portfolio:
equities_us:
holdings: [TSLA, NVDA, PLTR, GOOGL]
weights: [0.25, 0.30, 0.25, 0.20]
allocation_pct: 35
equities_jp:
holdings: [7203.T, 6758.T, 9984.T] # Toyota, Sony, SoftBank
weights: [0.40, 0.35, 0.25]
allocation_pct: 15
equities_fr:
holdings: [MC.PA, AI.PA, BN.PA] # LVMH, Air Liquide, Danone
weights: [0.50, 0.30, 0.20]
allocation_pct: 15
crypto:
holdings: [BTC, ETH, SOL]
weights: [0.50, 0.30, 0.20]
allocation_pct: 10
defi:
protocols: [aave_v3, morpho, eigenlayer]
allocation_pct: 10
prediction_markets:
platforms: [polymarket, kalshi]
allocation_pct: 15
risk_parameters:
var_confidence: [0.95, 0.99]
var_horizon: [1, 5, 21] # daily, weekly, monthly
method: historical_simulation
lookback_days: 756 # 3 years
stress_scenarios:
- name: "COVID Crash (Mar 2020)"
period: ["2020-02-20", "2020-03-23"]
- name: "FTX Collapse (Nov 2022)"
period: ["2022-11-06", "2022-11-14"]
- name: "SVB Crisis (Mar 2023)"
period: ["2023-03-08", "2023-03-15"]
Warren fetches price data, calculates returns, and runs the full risk suite:
pythonShow code
# Core risk calculation pipeline
from risk_engine import PortfolioRisk
risk = PortfolioRisk(config='portfolio-risk-config.yaml')
risk.fetch_data() # Pull all price histories
risk.calculate() # Run VaR, CVaR, Sharpe, Sortino, drawdown
risk.stress_test() # Apply historical scenarios
risk.generate_report() # Format for board pack
The output report:
View details
══════════════════════════════════════════════════════
PYRATZLABS PORTFOLIO RISK REPORT
Q1 2026 — Board Pack
══════════════════════════════════════════════════════
PORTFOLIO VALUE-AT-RISK
─────────────────────────────────────────────────────
Daily VaR Weekly VaR Monthly VaR
(95%) (99%) (99%)
─────────────────────────────────────────────────────
US Equities -1.8% -5.2% -11.4%
JP Equities -1.4% -4.1% -8.7%
FR Equities -1.2% -3.5% -7.3%
Crypto -4.6% -12.8% -26.1%
DeFi -5.1% -14.2% -28.8%
Pred. Markets -2.8% -7.4% -15.2%
─────────────────────────────────────────────────────
TOTAL PORTFOLIO -2.1% -6.0% -12.7%
─────────────────────────────────────────────────────
CVaR (Expected Shortfall):
Daily (95%): -3.4%
Weekly (99%): -9.8%
The Results
| Metric | Manual Process | Agent Process |
|---|---|---|
| Time to generate | 2-3 days | 2 minutes |
| Asset classes covered | Equities only (usually) | All 6 sub-portfolios |
| Stress scenarios | "We should add those" | 3 built-in, extensible |
| Update frequency | Quarterly | On-demand |
| Formatting for board | 2 hours in PowerPoint | Auto-generated |
| Sharpe ratio by book | Rarely calculated | Always included |
| CVaR included | Never (too complex in Excel) | Standard |
| Cost | CFO's sanity | Compute only |
The Sharpe ratio comparison was revealing: prediction markets (1.82) were our most risk-efficient book. Crypto (0.64) was the worst. That changed the allocation conversation.
Try It Yourself
- Sign up for Mr.Chief and install the
risk-metrics-calculationskill - Define your portfolio structure in YAML — supports any mix of asset classes
- Run
warren risk-report— fetches data, calculates everything, formats the output - Export to markdown (board pack) or pipe to the
table-imageskill for Telegram/Slack - Schedule weekly via cron for continuous risk monitoring
Risk management isn't about avoiding risk. It's about knowing exactly how much risk you're carrying — and being able to explain it in two minutes, not two days.
The board doesn't want to hear about your spreadsheet. They want to hear about their money.
Related case studies
Portfolio Manager
Regime Shifted to Contraction — Agent Moved to Defensive in 24 Hours
AI macro regime detection caught the expansion-to-contraction shift 3 weeks early. How our agent moved the portfolio to defensive positioning and preserved capital during a market drawdown.
Portfolio Manager
The Complete Financial Operating System — All 17 Skills in One Stack
How 17 AI finance skills work as one integrated system: data ingestion, analysis, execution, and reporting. 31 agents across 8 teams running a complete financial operating system on Mr.Chief.
Portfolio Manager
Earnings Preview Report for 5 Holdings — Before the Market Opens
How an AI agent generates earnings preview reports for NVDA, GOOGL, META, TSLA, and PLTR in 10 minutes — consensus estimates, implied moves, and key metrics delivered before Monday open.
Want results like these?
Start free with your own AI team. No credit card required.