All In One Finance Skills

All-in-One Finance Agent Skill Suite

Overview

Institutional-grade modular finance intelligence system covering equities, crypto, forex, commodities, fixed income, and derivatives. Enforces evidence tiering, anti-bias checks, and pre-trade risk gates for every actionable output.

Core principle: Every recommendation requires T1/T2 evidence backing. No T3-only signals. No skipping risk gates. No bias unchecked.

When to Use

digraph finance_trigger {
    "User mentions financial asset?" [shape=diamond];
    "Equity query?" [shape=diamond];
    "Crypto query?" [shape=diamond];
    "Forex/Commodity?" [shape=diamond];
    "Portfolio/Risk?" [shape=diamond];
    "Load equity modules" [shape=box];
    "Load crypto modules" [shape=box];
    "Load macro/forex modules" [shape=box];
    "Load risk-guardian" [shape=box];
    
    "User mentions financial asset?" -> "Equity query?" [label="yes"];
    "User mentions financial asset?" -> "Crypto query?" [label="yes"];
    "User mentions financial asset?" -> "Forex/Commodity?" [label="yes"];
    "User mentions financial asset?" -> "Portfolio/Risk?" [label="yes"];
    "Equity query?" -> "Load equity modules" [label="yes"];
    "Crypto query?" -> "Load crypto modules" [label="yes"];
    "Forex/Commodity?" -> "Load macro/forex modules" [label="yes"];
    "Portfolio/Risk?" -> "Load risk-guardian" [label="yes"];
}

Trigger keywords: $TICKER, BTC, ETH, EUR/USD, gold, oil, DCF, P/E, RSI, MACD, MVRV, NUPL, Fed, ECB, BoJ, carry trade, position size, Kelly, stop loss, drawdown, 10-K, earnings, whale, on-chain, sentiment, Fear & Greed, portfolio, hedge, correlation, beta, options, puts, calls, futures, contango, backwardation.

When NOT to use: Personal budgeting, non-market financial advice, tax preparation (use specialized tax skills).

Module Registry

RED FLAGS — STOP and Verify Evidence

• Recommendation with only T3 (opinion) sources
• Skipping Pre-Trade Risk Gate for "quick trades"
• Conviction >0.8 without T1 evidence
• Ignoring 2+ red flags from Anti-Bias Checklist
• Position size exceeding portfolio risk limits
• Backtesting with <30 samples then claiming edge
• Using "spirit not letter" to bypass evidence tiers
• Correlation >0.7 with existing positions but no reduction

All of these mean: STOP. Re-run gates. Gather T1/T2 evidence.

Evidence Standards (Non-Negotiable)

Rules:

1. No actionable recommendation (buy/sell/hedge) on T3-only evidence
2. Conviction score >0.5 requires ≥50% T1/T2 weighted evidence
3. Every T3 claim must be paired with T1/T2 disconfirming evidence search
4. Always disclose evidence composition in output

Anti-Bias Checklist (Run Before Every Recommendation)

6 Cognitive Traps

• Confirmation bias — Did I actively seek disconfirming evidence?
• Anchoring — Am I over-weighting first price/number seen?
• Recency bias — Am I ignoring 3+ year historical context?
• Herd mentality — Is consensus baked into my thesis without challenge?
• Sunk cost — Am I defending a prior call to avoid loss?
• Overconfidence — Is my conviction score calibrated to evidence quality?

10 Financial Red Flags (Scan Every Asset)

1. Revenue recognition changes / channel stuffing
2. Related-party transactions >5% revenue
3. Auditor changes or qualified opinions
4. Short interest spikes (>20% float in 30 days)
5. Insider selling clusters (3+ insiders in 90 days)
6. Covenant breaches or debt waivers
7. Whistleblower reports or SEC investigations
8. Off-balance-sheet SPVs or guarantees
9. Related-party leases or management contracts
10. Sudden CFO/audit committee turnover

Pre-Trade Risk Gate (5 Gates — All Must Pass)

Gate 1: Liquidity
  → Daily volume ≥ 10× position size?
  → Spread <0.5% (equities) / <0.1% (crypto large-cap)?
  → Market cap check: >$1B FULL | $100M–$1B REDUCED | <$100M SKIP

Gate 2: Correlation
  → 90d rolling correlation vs. portfolio <0.7?
  → Sector concentration <30% at full Kelly?
  → No >20% in single correlated cluster?

Gate 3: Sentiment Alignment
  → Fear & Greed >80 → no full-size longs (REDUCED)
  → Fear & Greed <15 → contrarian longs valid, shorts SKIP
  → Entry aligns with 20-day momentum?

Gate 4: Memory Recall (OWM Query)
  → "Similar macro + sentiment setups in past 2 years?"
  → 3+ negative outcomes → REDUCED
  → Behavioral drift detected → SKIP until review

Gate 5: Regulatory
  → Asset legal in user jurisdiction?
  → US: SEC/CFTC status, not unregistered security
  → EU: MiFID II appropriateness, ESMA limits
  → OFAC SDN list check (crypto wallet screening)

Output: FULL (proceed) | REDUCED (half size) | SKIP (block)

Query Classification (Step 1)

Before analysis, classify:

1. Asset Class: Equity / Crypto / Forex / Commodity / Fixed Income / Multi-Asset
2. Analysis Type: Fundamental / Technical / Sentiment / Forensic / Risk / Forecast
3. Complexity: Simple (1 module) / Composite (2–4 modules) / Full Framework (5+ modules)
4. User Profile: Retail (simplified) / Professional (full depth) / Quant (model-ready)

Then load only the relevant reference files identified.

Composition Workflows (Step 2)

Equity Deep Dive

fin-equity-fundamental → fin-equity-technical → fin-sentiment-engine → fin-risk-guardian

Crypto Cycle Positioning

fin-crypto-onchain + fin-macro-liquidity + fin-sentiment-engine → fin-risk-guardian

Crypto Bottom Signal

fin-crypto-onchain (NUPL <0) + fin-sentiment-engine (Fear <15) → conviction score

Forensic Alert Response

fin-crypto-forensic (drain/hack) → fin-news-aggregator → fin-risk-guardian (hedge)

Multi-Asset Hedge

fin-macro-liquidity + fin-forex-matrix + fin-commodity-cycle + fin-risk-guardian

Fixed Income Relative Value

fin-fixed-income + fin-macro-liquidity → credit spread analysis → fin-risk-guardian

Options Strategy

fin-options-derivatives + fin-equity-technical (timing) → fin-risk-guardian (Greeks check)

Structured Output (Step 3)

Every response follows this structure (adapt depth to complexity):

1. EXECUTIVE SUMMARY (3 bullets max)
   → Signal direction, conviction score, key catalyst

2. THESIS & VARIANT VIEW
   → Core bull/bear case
   → What could prove this wrong? (pre-mortem)

3. EVIDENCE MAP (tiered with URLs)
   → T1: [source] — [finding] — [URL]
   → T2: [source] — [finding] — [URL]
   → T3: [source] — [finding] — FLAG speculative

4. VALUATION / SCORE MATRIX
   → Quantified signals from each module
   → Fair value range (bear/base/bull)

5. RISK FACTORS
   → Bull / Base / Bear probabilities
   → Top 3 risks with mitigation

6. ACTION PLAN
   → Entry: $[price] or signal trigger
   → Exit: $[price] or thesis break
   → Stop: $[price] or [X%] drawdown
   → Size: [X%] portfolio (gate result)
   → Horizon: [days/weeks/months]

7. LOGIC CHAIN (for macro/event-driven)
   → Causal transmission diagram
   → Feedback loops and second-order effects

Quick query format (e.g., "what's BTC sentiment?"): Bullets 1 + 4 + 6 only.

Composure Under Pressure (Rationalization Defense)

Violating the letter of the rules is violating the spirit of the rules.

Compliance & Disclaimers

⚠️ All analyses are for educational and research purposes only. This is NOT financial advice. Past performance does not guarantee future results. Consult a licensed financial advisor before making investment decisions. Assets may be restricted in your jurisdiction. Options, futures, and crypto carry substantial risk of loss.

Append this disclaimer to ANY output containing:

• Specific buy/sell/hedge recommendations
• Position sizing suggestions
• Portfolio allocation advice
• Options/futures strategies

Reference File Loading Strategy

DO NOT load all files at once. Load on-demand:

• Equity analysis: equity-fundamental.md + equity-technical.md
• Crypto analysis: crypto-onchain.md + crypto-forensic.md
• Macro research: macro-liquidity.md + forex-matrix.md + commodity-cycle.md
• Risk assessment: risk-guardian.md + fixed-income.md
• Derivatives: options-derivatives.md + algo-execution.md
• Reporting: report-orchestrator.md + predictor-kronos.md
• Data/News: news-aggregator.md + sentiment-engine.md
• Audit/Review: memory-protocol.md

Quick Reference: Asset Class Decision Tree

digraph asset_class {
    "Ticker with $?" [shape=diamond];
    "Crypto keywords?" [shape=diamond];
    "Forex pair format?" [shape=diamond];
    "Commodity name?" [shape=diamond];
    "Bond/fixed income?" [shape=diamond];
    "Equity" [shape=box];
    "Crypto" [shape=box];
    "Forex" [shape=box];
    "Commodity" [shape=box];
    "Fixed Income" [shape=box];
    "Multi-Asset" [shape=box];
    
    "Ticker with $?" -> "Crypto keywords?" [label="no"];
    "Ticker with $?" -> "Equity" [label="yes"];
    "Crypto keywords?" -> "Forex pair format?" [label="no"];
    "Crypto keywords?" -> "Crypto" [label="yes"];
    "Forex pair format?" -> "Commodity name?" [label="no"];
    "Forex pair format?" -> "Forex" [label="yes"];
    "Commodity name?" -> "Bond/fixed income?" [label="no"];
    "Commodity name?" -> "Commodity" [shape=box] [label="yes"];
    "Bond/fixed income?" -> "Multi-Asset" [label="no"];
    "Bond/fixed income?" -> "Fixed Income" [label="yes"];
}