OpenClaw Equity Research

Goal

Produce decision-ready equity research that combines market data, company context, catalysts, valuation framing, risk checks, and an explicit evidence trail.

This skill is inspired by:

• OpenBB as the data-platform pattern: connect data once, consume it in reports, terminals, dashboards, or agents.
• Agentic stock research systems as the workflow pattern: split work into stock finder, market data, news/catalyst, and recommendation synthesis stages.

Use Cases

• Single ticker research memo: AAPL, TSLA, NVDA, RKLB, etc.
• Watchlist triage: compare multiple tickers and rank research priority.
• Company deep dive: business model, market structure, financial quality, catalysts, valuation, risks.
• Trading-oriented note: technical setup, levels, momentum, stop/risk framing.
• Long-term investor note: moat, growth, margins, capital allocation, valuation scenario.
• OpenClaw workflow design for analyst teams, research terminals, or agentic investment workspaces.

Hard Rules

• Do not present output as financial advice. Use research language, not instructions to trade.
• Separate facts, estimates, and judgment.
• Cite or name sources for all nontrivial claims when sources are available.
• Prefer fresh data. For current prices, news, estimates, filings, and analyst changes, browse or use data APIs unless the user explicitly forbids it.
• If data is stale, missing, or provider-limited, say so in the memo.
• Do not fabricate financial metrics, target prices, filings, analyst ratings, or news.
• For high-stakes recommendations, include bear case, key downside risks, and what would falsify the thesis.

Quick Start

From this skill directory:

python3 scripts/equity_research.py AAPL --out reports
python3 scripts/equity_research.py TSLA NVDA RKLB --mode watchlist --out reports
python3 scripts/equity_research.py --template AAPL --out reports

The script writes:

• {ticker}-equity-research.md
• {ticker}-equity-research.json
• or watchlist-equity-research.md for multi-ticker triage

Research Workflow

1. Clarify scope only when necessary: ticker(s), market, time horizon, user intent, and risk tolerance.
2. Gather data:
   • price history, volume, technical posture
   • company profile, sector, market cap
   • recent news/catalysts
   • fundamentals and valuation metrics when available
   • filings/transcripts/earnings if the user needs a deep dive
3. Run the stage model:
   • Finder/Triage: why this ticker is in scope and what makes it worth research time.
   • Market Data: price, trend, liquidity, volatility, technical levels.
   • News/Catalyst: recent events, sentiment, near-term calendar.
   • Fundamental/Valuation: revenue quality, margin trend, cash generation, balance sheet, multiples or scenario frame.
   • Synthesis: base case, bull case, bear case, key risks, monitoring points.
4. Produce a memo with an explicit evidence table and confidence level.
5. If the user asks for an action label, use Research View: Bullish / Neutral / Bearish, not personalized investment advice.

Output Shape

Use this memo structure unless the user requests a different format:

# {TICKER} Equity Research Memo

## Snapshot
- Research view:
- Time horizon:
- Current price / market cap:
- Data timestamp:
- Confidence:

## Thesis

## Evidence
| Area | Evidence | Source / timestamp | Interpretation |

## Market Data And Technical Setup

## Company And Fundamentals

## Catalysts

## Valuation Frame
- Base case:
- Bull case:
- Bear case:
- Key assumptions:

## Risks And Falsification

## Monitoring Checklist

## Research Limits

When To Load References

• Read references/research-framework.md before writing a full memo, building an agent workflow, or modifying the script.
• Read references/data-sources.md when choosing between OpenBB, yfinance, SEC/filings, news, or web sources.
• Read references/report-rubric.md when the user asks for institutional quality or review-ready output.

Script Notes

scripts/equity_research.py is intentionally lightweight:

• OpenBB-style design: one script produces reusable JSON + markdown artifacts.
• yfinance-first runtime because it is already available in many OpenClaw environments.
• OpenBB can be added later as a provider layer without changing the memo contract.
• The script's output is a research starting point; the agent should still add judgment, source checks, and user-specific context when requested.