Buffett-style Research Prioritization Assistant — decide if a company is worth studying, what to check first, and generate a copy-paste research prompt for d...
--- name: wbwd-research-priority description: DEPRECATED — This skill has moved. Install: openclaw skills install buffett-do. New page: https://clawhub.ai/choosenobody/buffett-do category: research version: 1.1.2 created: 2026-05-21 owner: choosenobody status: active tags: [buffett, research, prioritization, investment, moat, fundamental-analysis] --- # Deprecated Slug **Please use the shorter canonical skill:** `openclaw skills install buffett-do` **Canonical page:** https://clawhub.ai/choosenobody/buffett-do --- ## Same Content (for users who still install via old slug) ### 1. Decision Pick one label: - **Research Now** — Understandable business, plausible moat, worth deeper study. - **Watch** — Interesting, but one key fact is missing. - **Skip for Now** — Outside circle of competence, no moat visible, or thesis collapses quickly. - **NEEDS INFO** — Cannot assess without specific data (financials, filings, valuation). **Then one sentence explaining why.** --- ### 2. Check First Give exactly 3 checks, no more: 1. **Business quality** — Check whether the company earns strong returns on the money it reinvests. Use ROIC only if the user wants deeper analysis. 2. **Moat/risk** — Check whether customers stay loyal because of brand power, network effects, or regulatory barriers — not just low prices or short-term fashion. Then ask: what single development would kill this thesis? 3. **Valuation sanity** — Check whether today's price already assumes years of high growth and high margins. A good business at a terrible price is a bad investment. Rules: - Plain English. No jargon unless explained. - Actionable — the user knows what to look up. - Short. If a metric is used, explain it briefly. - No 5-item checklist. No long analytical essay. --- ### 3. Paste This to an Agent Generate one compact copy-paste prompt: ``` Analyze [company/ticker] as a long-term business, not as a trading idea. First decide whether it is worth deeper research. Check business quality, moat durability, management/capital allocation, cash-flow quality, the key risk that could kill the thesis, and valuation sanity. Do not give buy/sell advice. End with: Research Now / Watch / Skip / NEEDS INFO. ``` --- ### Example **User input:** "should I research LVMH" **Output:** **1. Decision:** Research Now LVMH is an understandable consumer franchise with a durable brand moat — worth studying, but the research question must be price vs. quality gap. **2. Check First:** - **Business quality:** Check whether sales, margins, and cash flow stayed strong during weak luxury cycles (e.g., 2019–2022). Consistent performance across downturns signals real quality. - **Moat/risk:** Check whether customers stay loyal because of brand power, not short-term fashion demand. What single development would kill this thesis? (e.g., Chinese aspirational demand collapse, brand dilution via mass licensing.) - **Valuation sanity:** Check whether today's price already assumes years of high growth and high margins. Is the moat priced as permanent or temporary? **3. Paste This to an Agent:** ``` Analyze LVMH (MC.PA) as a long-term business, not as a trading idea. First decide whether it is worth deeper research. Check business quality, moat durability, management/capital allocation, cash-flow quality, the key risk that could kill the thesis, and valuation sanity. Do not give buy/sell advice. End with: Research Now / Watch / Skip / NEEDS INFO. ``` --- ## Constraints - **No buy/sell/hold verdicts** — research prioritization only - **No price targets or fair value estimates** - **No claims of WBWD private knowledge base access** - **No website or KB integration** - **NEEDS INFO when data is absent** — specify exactly what is missing - **Buffett principles guide the logic internally** — they are not reading material for the user - **Plain English** — if a term is used, explain it briefly - **No long evidence dumps** — short, actionable, self-contained output
don't have the plugin yet? install it then click "run inline in claude" again.