public-equity-research

Perform recommendation-grade research on a named listed company using exact security identity, point-in-time primary filings, business/segment economics, earnings quality, balance sheet, cash flow, di

SKILL.md

GitHub

Public Equity Research

Own deep research on a named public company and its exact listed security. Use the user's language and distinguish business quality, security valuation/readiness and portfolio action.

Read references/equity-research-protocol.md. If the installed $public-equity-investing plugin is available, use its narrow earnings, valuation, scenario or thesis workflows to deepen this protocol; if unavailable, complete the same evidence categories directly from current primary filings and authoritative sources.

Workflow

  1. Resolve identity and cutoff. Confirm legal issuer, ticker, exchange, security/share class, currency, ADR/depositary ratio, fiscal calendar and data cutoff. Separate company value from the exact security claim.
  2. Build the source ledger. Start with current annual/interim filings, earnings release, notes, investor presentation only as management claims, exchange/regulator disclosures and official share/capital records. Apply $finance-evidence-guard to decisive figures.
  3. Understand the business. Map segments, customers, geography, revenue drivers, unit economics, pricing, cyclicality, competitive position, capital intensity, regulation and key dependencies.
  4. Normalize financials. Reconcile accounting basis, currency, consolidation, inflation treatment, fiscal periods, acquisitions/disposals, one-offs, stock compensation, leases, capitalized costs and continuing operations. Build revenue, margin, cash conversion, returns on capital, leverage and dilution trends.
  5. Test earnings quality and balance sheet. Compare profit with operating/free cash flow, working capital, capex, receivables/inventory, provisions, related parties, refinancing, covenants, pension/off-balance-sheet obligations and contingent liabilities.
  6. Estimate what is priced in. Use at least two compatible valuation frames—such as historical/peer multiples, reverse DCF, unit economics or asset value. Make assumptions visible, use scenario ranges and show the breakpoints that explain the current price.
  7. Define thesis and variant. State what consensus/price appears to assume, why the evidence differs, dated catalysts/recognition path, KPIs, strongest contradiction and hard kill criteria. Route persistent monitoring to $investment-thesis-tracker.
  8. Add market context. Use $technical-quant-analysis for timing, $market-regime-analysis for cross-asset context, and $probabilistic-market-forecast for outcome ranges. Use $turkey-markets-analysis for BIST/KAP/TMS 29/Türkiye rules.
  9. Challenge and decide. Compare a relevant peer/alternative and doing nothing, then run $investment-red-team. Use $portfolio-risk-and-sizing for allocation and $pre-trade-investment-gate if action is imminent.

Hard rules

  • An open-ended “best stock” claim must start with $equity-opportunity-funnel; a named-company report cannot establish a market-wide winner.
  • Do not treat management guidance, analyst targets or investor presentations as independent proof.
  • Do not mix periods, currencies, consolidated bases, reported/adjusted metrics or nominal/real figures silently.
  • Do not hide dilution, leverage, cyclicality, governance, liquidity or valuation sensitivity behind a narrative moat.
  • Do not convert a precise spreadsheet output into precise confidence; ranges must reflect model and evidence uncertainty.

Output

Lead with action or research verdict, horizon, current price/time and confidence. Then show identity, business/segments, normalized financial trends, earnings quality/balance sheet, valuation and what is priced in, bear/base/bull cases, catalysts, contradictions/kill criteria, peer/alternative, technical/regime context when material, portfolio/execution conditions and source ledger.