Ai Trading Consortium
by @ericn26-star
AI-powered hedge fund that combines multi-expert trading strategies with comprehensive information gathering. Analyzes stocks through systematic data collect...
clawhub install ai-trading-consortiumπ About This Skill
name: ai-trading-consortium description: "AI-powered hedge fund that combines multi-expert trading strategies with comprehensive information gathering. Analyzes stocks through systematic data collection, analyst reports, bull/bear debates, and a council of legendary investor personas (Buffett, Graham, Lynch, Burry, Munger, Wood, Druckenmiller, Marks). Produces executive summary slides. Trigger keywords: stock analysis, trading decision, investment analysis, market evaluation, buy/sell/hold recommendation."
AI Trading Consortium
Overview
An AI-powered hedge fund system that combines sophisticated information gathering with multi-expert trading wisdom. It coordinates a full analysis pipeline: data collection, four-analyst assessment, structured bull/bear debate, an 8-member legendary investor council, risk management, portfolio decision, and executive summary slide generation (PDF + PPTX).
Important: Never provide financial advice β always clarify this is for educational/research purposes only.
Workflow
Phase 0: Stock Identification (if needed)
If the user provides a company name, partial name, or ambiguous query instead of a ticker symbol:
1. Search to identify the correct stock ticker and exchange 2. Confirm the company identity (full name, ticker, exchange, sector) 3. If multiple matches exist, clarify with the user or select the most relevant
Examples:
Output for symbol lookup:
{
"query": "user's original query",
"identified_stock": {
"company_name": "Full Legal Name Inc.",
"ticker": "TICK",
"exchange": "NASDAQ",
"sector": "Technology",
"confidence": "high/medium/low"
},
"alternatives": [
{"ticker": "OTHER", "reason": "Similar name..."}
]
}
Phase 1: Information Gathering
1. Collect comprehensive market data from multiple sources (see Data Sources section) 2. Run all four analyst assessments sequentially: - Fundamentals Analyst β Financial health, valuation metrics - Technical Analyst β Price patterns, indicators - Sentiment Analyst β Social media, market sentiment - News Analyst β Breaking news, macro events
Each analyst produces a structured report (see Analyst Frameworks below).
Phase 2: Research & Debate
3. Construct both bull and bear cases from the analyst reports 4. Conduct 2-3 rounds of structured debate (Opening Arguments β Rebuttals β Final Arguments) 5. Synthesize debate conclusions with a winner, scoring breakdown, and key takeaways
Phase 3: Expert Council Review
6. Channel each of the 8 legendary investor personas to evaluate the opportunity 7. Gather diverse perspectives β each investor votes BUY/HOLD/SELL with conviction (1-10) and reasoning 8. Weight opinions based on relevance to the specific opportunity
Phase 4: Risk & Execution
9. Evaluate risk: position sizing, stop-losses, portfolio impact, VaR 10. Make final trading decision: synthesize all inputs into an actionable recommendation
Phase 5: Executive Summary Slides
11. Create 3 PPT Summary Slides (see Slide Specification below): - Slide 1: Process Overview & Data Collected - Slide 2: Expert Council Voting Results (hero visual with 8 investor votes) - Slide 3: Final Decision & Action Plan 12. Compile slides into PDF and PPTX deliverables
Output Format
For each analysis, provide all of the following:
1. Executive Summary: Key findings and recommendation 2. Analyst Reports: Summary of each analyst's findings 3. Debate Summary: Bull vs Bear key arguments 4. Expert Council Votes: How each investor persona voted and why 5. Risk Assessment: Position size, stop-loss, risk/reward ratio 6. Final Decision: BUY / SELL / HOLD with conviction level (1-10) 7. Three Summary Slides: Visual presentation of the entire analysis
Data Sources
Configure data gathering from multiple sources:
| Category | Sources | |----------|---------| | Stock Price Data | Yahoo Finance (yfinance) β Primary for OHLCV; Alpha Vantage β Validation | | Fundamental Data | Alpha Vantage, SEC filings (10-K, 10-Q, 8-K) | | Technical Indicators | Yahoo Finance, Alpha Vantage | | News Data | Alpha Vantage, Google News, financial news APIs | | Sentiment Data | Social media APIs (Reddit, Twitter/X, StockTwits) |
Data Gathering Output Structure
{
"ticker": "SYMBOL",
"timestamp": "ISO datetime",
"price_data": {
"current_price": 0.00,
"52_week_high": 0.00,
"52_week_low": 0.00,
"volume": 0,
"avg_volume": 0
},
"fundamentals": {
"market_cap": 0,
"pe_ratio": 0.00,
"pb_ratio": 0.00,
"revenue_growth": 0.00,
"earnings_growth": 0.00
},
"technicals": {
"rsi_14": 0.00,
"macd_signal": "bullish/bearish",
"trend": "uptrend/downtrend/sideways"
},
"news_summary": "Recent news highlights",
"sentiment_score": 0.00
}
Data Quality Guidelines
Analyst Frameworks
Fundamentals Analyst
Evaluate companies based on financial health, competitive position, and intrinsic value.
#### Financial Health Assessment
1. Profitability - Gross margin, operating margin, net margin - Return on Equity (ROE), Return on Assets (ROA) - Return on Invested Capital (ROIC)
2. Growth Metrics - Revenue growth (YoY, 3-year CAGR) - Earnings growth - Free cash flow growth
3. Balance Sheet Strength - Current ratio, quick ratio - Debt/equity ratio - Interest coverage ratio - Cash position
4. Cash Flow Quality - Operating cash flow vs net income - Free cash flow generation - Capital allocation efficiency
#### Valuation Methods
1. Discounted Cash Flow (DCF) β Project 5-10 year cash flows, apply WACC, calculate terminal value, derive intrinsic value per share 2. Comparable Company Analysis β Identify peer group, compare P/E, EV/EBITDA, P/S ratios, apply appropriate multiples 3. Dividend Discount Model (DDM) β For dividend-paying stocks; Gordon Growth Model or multi-stage DDM 4. Asset-Based Valuation β Book value analysis, Net Asset Value (NAV)
#### Fundamentals Report Format
# Fundamentals Analysis Report: [TICKER]Financial Health Score: [1-10]
Profitability Analysis
[Key findings] Growth Assessment
[Key findings] Balance Sheet Strength
[Key findings] Cash Flow Quality
[Key findings] Valuation Summary
| Method | Intrinsic Value | Upside/Downside |
|--------|-----------------|-----------------|
| DCF | $XX.XX | +/-XX% |
| Comps | $XX.XX | +/-XX% |Signal: [BULLISH/BEARISH/NEUTRAL]
Confidence: [1-10]
Key Risks
[Risk factors]
Technical Analyst
Analyze price patterns, volume, and technical indicators to forecast price movements and identify entry/exit points.
#### Analysis Framework
Trend Analysis:
Momentum Indicators:
Volume Analysis:
Chart Patterns:
Support & Resistance:
#### Technical Report Format
# Technical Analysis Report: [TICKER]Trend Summary
Primary Trend: [UPTREND/DOWNTREND/SIDEWAYS]
Short-term: [description]
Medium-term: [description] Key Indicators
| Indicator | Value | Signal |
|-----------|-------|--------|
| RSI (14) | XX.X | Bullish/Bearish/Neutral |
| MACD | XX.X | Bullish/Bearish/Neutral |
| 50 SMA | $XX.X | Above/Below |
| 200 SMA | $XX.X | Above/Below |Support & Resistance
Key Support: $XX.XX, $XX.XX
Key Resistance: $XX.XX, $XX.XX Pattern Recognition
[Identified patterns and implications] Signal: [BULLISH/BEARISH/NEUTRAL]
Confidence: [1-10]
Entry/Exit Levels
Suggested Entry: $XX.XX
Stop Loss: $XX.XX
Target 1: $XX.XX
Target 2: $XX.XX
Sentiment Analyst
Analyze market sentiment from various sources to understand investor psychology and crowd behavior.
#### Sentiment Sources
Social Media:
Investor Behavior:
Market Indicators:
#### Sentiment Scoring
Composite sentiment score from -1 to +1:
| Range | Interpretation | |-------|---------------| | -1.0 to -0.5 | Extremely Bearish | | -0.5 to -0.2 | Bearish | | -0.2 to +0.2 | Neutral | | +0.2 to +0.5 | Bullish | | +0.5 to +1.0 | Extremely Bullish |
#### Sentiment Report Format
# Sentiment Analysis Report: [TICKER]Overall Sentiment Score: [X.XX] ([BULLISH/BEARISH/NEUTRAL])
Social Media Sentiment
| Source | Score | Trend | Notable |
|--------|-------|-------|---------|
| Reddit | X.XX | β/β | [notes] |
| Twitter| X.XX | β/β | [notes] |Investor Behavior
Put/Call Ratio: X.XX ([interpretation])
Short Interest: X.X% ([interpretation])
Insider Activity: [Net buying/selling] Institutional Sentiment
Recent 13F changes: [summary]
Notable positions: [key funds] Contrarian Signals
[Any extreme readings that suggest reversal] Signal: [BULLISH/BEARISH/NEUTRAL]
Confidence: [1-10]
Key Observations
[Notable sentiment shifts or anomalies]
News Analyst
Monitor and analyze news, events, and macroeconomic factors impacting stock prices.
#### Coverage Areas
Company-Specific News:
Industry & Sector:
Macroeconomic Events:
#### Impact Assessment
Rate each news item:
#### News Report Format
# News Analysis Report: [TICKER]Breaking News Summary
| Date | Headline | Impact | Direction |
|------|----------|--------|-----------|
| MM/DD | [news] | High/Med/Low | +/-/= |Company News
Recent Developments
[Key news items with analysis] Upcoming Events
[Earnings date, ex-dividend, etc.] Industry Context
[Relevant industry news] Macro Environment
[Relevant macro factors] News-Driven Catalysts
Positive Catalysts
[List with timing] Negative Risks
[List with timing] Signal: [BULLISH/BEARISH/NEUTRAL]
Confidence: [1-10]
Key Takeaways
[Summary of most important news impacts]
Debate Framework
Take all four analyst reports and conduct a structured investment debate.
Round 1: Opening Arguments
BULL CASE β Build arguments around:
BEAR CASE β Build arguments around:
Round 2: Rebuttals
Round 3: Final Arguments
Debate Rules
1. Be Specific: Use actual data points from analyst reports 2. Be Fair: Acknowledge valid opposing arguments 3. Quantify: Provide price targets and percentages 4. No Fabrication: Only use information from provided reports
Debate Output Format
# INVESTMENT DEBATE: [TICKER]ROUND 1: OPENING ARGUMENTS
π BULL CASE
Thesis: [One sentence]Key Arguments:
1. [Argument 1]
- Evidence: [data from reports]
- Implication: [why this matters]
2. [Argument 2]
- Evidence: [data from reports]
- Implication: [why this matters]
3. [Argument 3]
- Evidence: [data from reports]
- Implication: [why this matters]
Upside Target: $XX.XX (+XX%)
π» BEAR CASE
Thesis: [One sentence]Key Arguments:
1. [Argument 1]
- Evidence: [data from reports]
- Implication: [why this matters]
2. [Argument 2]
- Evidence: [data from reports]
- Implication: [why this matters]
3. [Argument 3]
- Evidence: [data from reports]
- Implication: [why this matters]
Downside Target: $XX.XX (-XX%)
ROUND 2: REBUTTALS
π Bull Rebuts Bear
| Bear Argument | Bull Counter |
|---------------|--------------|
| [point] | [rebuttal] |
| [point] | [rebuttal] |π» Bear Rebuts Bull
| Bull Argument | Bear Counter |
|---------------|--------------|
| [point] | [rebuttal] |
| [point] | [rebuttal] |
ROUND 3: CLOSING ARGUMENTS
π Bull's Final Statement
[2-3 sentences summarizing strongest case]π» Bear's Final Statement
[2-3 sentences summarizing strongest case]
DEBATE VERDICT
Winner: [BULL / BEAR]
Margin: [Decisive / Narrow / Too Close to Call]
Confidence: [1-10]
Scoring Breakdown
| Category | Bull Score | Bear Score |
|----------|------------|------------|
| Evidence Quality | X/10 | X/10 |
| Logic & Reasoning | X/10 | X/10 |
| Risk Assessment | X/10 | X/10 |
| Total | XX/30 | XX/30 |Key Takeaways
Strongest Bull Point: [summary]
Strongest Bear Point: [summary]
Unresolved Uncertainty: [what remains unclear] Recommendation to Portfolio Manager
[Summary of debate outcome and suggested weighting of bull vs bear perspectives]
Multi-Expert Investment Council
Channel 8 legendary investors, each with distinct investment philosophies.
Council Members
#### 1. Warren Buffett (Value/Quality) Philosophy: "Buy wonderful companies at fair prices"
#### 2. Benjamin Graham (Deep Value) Philosophy: "Margin of safety is paramount"
#### 3. Peter Lynch (Growth at Reasonable Price) Philosophy: "Invest in what you know"
#### 4. Michael Burry (Contrarian Deep Value) Philosophy: "Bet against the crowd when they're wrong"
#### 5. Charlie Munger (Quality at Fair Price) Philosophy: "All intelligent investing is value investing"
#### 6. Cathie Wood (Disruptive Innovation) Philosophy: "Invest in disruptive innovation"
#### 7. Stanley Druckenmiller (Macro/Momentum) Philosophy: "Soros taught me to bet big when you're right"
#### 8. Howard Marks (Risk/Cycles) Philosophy: "Risk control is paramount"
Council Output Format
# INVESTMENT COUNCIL VERDICT: [TICKER]Council Summary
| Investor | Verdict | Conviction | Key Reasoning |
|----------|---------|------------|---------------|
| Buffett | BUY/HOLD/SELL | 1-10 | [one line] |
| Graham | BUY/HOLD/SELL | 1-10 | [one line] |
| Lynch | BUY/HOLD/SELL | 1-10 | [one line] |
| Burry | BUY/HOLD/SELL | 1-10 | [one line] |
| Munger | BUY/HOLD/SELL | 1-10 | [one line] |
| Wood | BUY/HOLD/SELL | 1-10 | [one line] |
| Druckenmiller | BUY/HOLD/SELL | 1-10 | [one line] |
| Marks | BUY/HOLD/SELL | 1-10 | [one line] |Individual Perspectives
Warren Buffett's View
[2-3 paragraph analysis in Buffett's style]Benjamin Graham's View
[2-3 paragraph analysis in Graham's style]Peter Lynch's View
[2-3 paragraph analysis in Lynch's style]Michael Burry's View
[2-3 paragraph analysis in Burry's style]Charlie Munger's View
[2-3 paragraph analysis in Munger's style]Cathie Wood's View
[2-3 paragraph analysis in Wood's style]Stanley Druckenmiller's View
[2-3 paragraph analysis in Druckenmiller's style]Howard Marks's View
[2-3 paragraph analysis in Marks's style]Consensus Analysis
Bulls: [X] investors (names)
Bears: [X] investors (names)
Neutral: [X] investors (names) Weighted Council Decision
Aggregate Score: X.X / 10
Recommendation: [BUY/HOLD/SELL]
Key Agreement Areas: [common themes]
Key Disagreements: [divergent views]
Risk Management
Evaluate risks, calculate position sizes, and ensure the portfolio stays within defined risk parameters.
Position Sizing
Calculate optimal position size based on:
Risk Metrics
Per-Position:
Portfolio-Level:
Stop-Loss Framework
Hard Risk Limits
| Limit | Threshold | |-------|-----------| | Maximum single position | 10% of portfolio | | Maximum sector exposure | 25% | | Maximum drawdown tolerance | 15% | | Maximum portfolio VaR (95%) | X% |
Soft Limits (requiring justification)
Risk Assessment Categories
| Risk Category | Low | Medium | High | Critical | |---------------|-----|--------|------|----------| | Volatility | <20%| 20-40% | 40-60% | >60% | | Liquidity | >10M| 1-10M | 100K-1M| <100K | | Position Size | <2% | 2-5% | 5-10% | >10% | | Correlation | <0.3| 0.3-0.5| 0.5-0.7| >0.7 |
Risk Report Format
# RISK ASSESSMENT: [TICKER]Position Risk Profile
| Metric | Value | Rating |
|--------|-------|--------|
| Volatility (30d) | XX% | Low/Med/High |
| Beta | X.XX | Low/Med/High |
| Average Volume | $XXM | Adequate/Concern |
| VaR (95%, 1d) | -X.X% | Low/Med/High |Recommended Position Sizing
Max Position Size: $XXX,XXX (X% of portfolio)
Sizing Method: [Kelly/Volatility-adjusted/Fixed]
Rationale: [explanation] Stop-Loss Recommendations
Initial Stop: $XX.XX (-X%)
Trailing Stop: X ATR or X%
Max Loss per Trade: $X,XXX (X% of portfolio) Portfolio Impact Analysis
Current portfolio correlation: X.XX
Sector exposure after trade: X%
New portfolio VaR: X% Risk Flags
π’ [Low risk items]
π‘ [Medium risk items]
π΄ [High risk items requiring attention] Risk-Adjusted Recommendation
Risk/Reward Ratio: X:1
Approval Status: APPROVED / APPROVED WITH CONDITIONS / REJECTED
Conditions: [if applicable]
Portfolio Manager Decision Framework
Decision Criteria
Strong Buy Signals:
Strong Sell Signals:
Input Weighting
| Category | Weight | Notes | |----------|--------|-------| | Fundamentals | 25% | [key point] | | Technicals | 20% | [key point] | | Sentiment | 15% | [key point] | | News | 15% | [key point] | | Expert Council | 25% | [X/8 bullish] |
Portfolio Considerations
Rebalancing Triggers
Final Decision Output Format
# TRADING DECISION: [TICKER]Executive Summary
[2-3 sentence summary of the decision and key reasoning]Decision: [STRONG BUY / BUY / HOLD / SELL / STRONG SELL]
Conviction: [1-10]
Analysis Summary
| Category | Signal | Weight | Notes |
|----------|--------|--------|-------|
| Fundamentals | Bull/Bear/Neutral | 25% | [key point] |
| Technicals | Bull/Bear/Neutral | 20% | [key point] |
| Sentiment | Bull/Bear/Neutral | 15% | [key point] |
| News | Bull/Bear/Neutral | 15% | [key point] |
| Expert Council | Bull/Bear/Neutral | 25% | [X/8 bullish] |Research Debate Outcome
Winner: [BULL/BEAR]
Key Arguments: [summary] Trade Order
| Field | Value |
|-------|-------|
| Action | BUY / SELL |
| Ticker | [TICKER] |
| Quantity | [shares] |
| Order Type | LIMIT / MARKET |
| Limit Price | $XX.XX |
| Stop Loss | $XX.XX (-X%) |
| Target 1 | $XX.XX (+X%) |
| Target 2 | $XX.XX (+X%) |
| Time Horizon | [days/weeks/months] |Risk Management
Position Size: X% of portfolio
Max Loss: $X,XXX (X% of portfolio)
Risk/Reward: X:1 Thesis & Exit Criteria
Investment Thesis:
[Clear statement of why we're making this trade]Exit Triggers:
[ ] Target price reached
[ ] Stop loss hit
[ ] Thesis broken: [specific conditions]
[ ] Time stop: [date if applicable] Disclaimers
This is for educational/research purposes only. Not financial advice. Past performance does not guarantee future results. Always do your own research.
Executive Summary Slides Specification
Create 3 visually polished slides that feel like a Premium Financial Journal or High-End Strategic Report.
Design Aesthetic Guidelines (CRITICAL)
Slide Generation Process
Use image generation tools to create slides:
1. Write content_script.md as pure information architecture before generating
2. Use gen_images for the first slide (create from scratch)
3. Use edit_images with base_image_file pointing to previous slide for subsequent slides (ensures visual consistency)
4. Generate sequentially, not in parallel
5. Compile into PDF (150 DPI, 95% quality) and PPTX upon completion
6. DO NOT generate summary documents or design descriptions β deliver only the visual files
Slide 1: Process Overview & Canvas
*Design concept: Minimalist infographic track.*
Slide 2: The Expert Council (HERO SLIDE)
*Design concept: "The Council Table" β Visual arrangement of investor personas, NOT a standard Excel-style table.*
Create a visual card or portrait layout for the 8 investors. Do NOT use a standard table.
Data to Visualize:
| Investor | Vote | Conviction | Key Reasoning (Short) | |----------|------|------------|----------------------| | Warren Buffett | [Vote] | X/10 | [Reasoning] | | Benjamin Graham | [Vote] | X/10 | [Reasoning] | | Peter Lynch | [Vote] | X/10 | [Reasoning] | | Michael Burry | [Vote] | X/10 | [Reasoning] | | Charlie Munger | [Vote] | X/10 | [Reasoning] | | Cathie Wood | [Vote] | X/10 | [Reasoning] | | Stanley Druckenmiller | [Vote] | X/10 | [Reasoning] | | Howard Marks | [Vote] | X/10 | [Reasoning] |
Visual concept options (choose the most impactful):
Design emphasis: This should look like an exciting "moment of truth" reveal, not a boring table.
Slide 3: The Verdict & Execution Art
*Design concept: Clean, asymmetric editorial layout.*
Slide Visual Design Principles
When constructing image generation prompts for each slide:
1. Visualization type: Prioritize diagram forms over text-dominated presentations (cutaway views, flowcharts, annotated structure diagrams, relationship maps, timeline overlays). Avoid "parallel cards/grid display/multi-column" and text-heavy traditional layouts 2. Information hierarchy: Primary vs secondary info distinguished through visual hierarchy (size, position, contrast), not flat listing 3. Composition: Asymmetric layout, diagonal momentum, or other approaches that break rigid symmetry 4. Density: Information hierarchy clarity takes priority over quantity; appropriate whitespace serves readability 5. Layout independence: Each slide's visualization type chosen based on its own content, not copying previous slide 6. Style consistency: All 3 slides share cohesive visual identity (color palette, typography, layout DNA) while each having its own structure