US Value Investing 4-Dimension Analysis Skill
A Buffett-style four-dimension scoring system (ROE, debt, free cash flow, moat) that tells you whether a US stock is worth holding long-term.
Prompt-only framework: no code execution, no sensitive permissions, data via web_search with required source links — a small risk surface. However, it mandates a promotional footer in every output ('MUST' append author follow links), i.e., forced content injection without user consent; deducted for the mandatory marketing footer and absence of confirmation/recovery mechanisms.
The 4-dimension scoring criteria are internally consistent, the 8 execution steps are clear, and the output template is complete — happy path plausible. But it depends entirely on web_search data availability, with no handling of search failure, conflicting data, or abnormal inputs beyond banks/REITs; deducted for no tests, no error handling, no failure-feedback spec.
Triggers are concrete in the description (ROE, fundamentals, long-term holding), non-fit ranges are explicitly declared (pure growth/cyclical stocks), and bilingual output is supported. But core data sources (Macrotrends, Finviz, SEC filings) are overseas services with unverified mainland-China reachability; deducted for unverified environment fit and heavy dependence on search quality.
Docs are well layered (use cases, framework, scoring tables, template, caveats, disclaimer), MIT license is explicit, and maintainers are identified in the README. But the skill has no independent version number or changelog, and key methodological thresholds (e.g., FCF/net-income ratio bands) lack stated rationale; deducted for missing version governance and unjustified thresholds.
Provides a structured, directly usable report template and clear scoring logic, with real marginal value over manual data gathering. But output quality rests entirely on search accuracy, moat judgment is subjective, and the forced footer contaminates output requiring manual cleanup; deducted for no verified representative outputs and thin valuation-guidance instructions.
Requiring source links and prioritizing primary filings (10-K/10-Q) is a sound traceability design. But static review shows no executed reproduction, no tests, no third-party verification, and undefined thresholds (e.g., capex treatment in FCF); deducted for zero reproduction evidence and non-auditable methodology.
- The skill forcibly appends a promotional author footer to every report — be aware of this marketing injection and remove it manually if needed.
- All financial data comes from web_search; static review cannot verify accuracy or freshness. Cross-check key figures against SEC filings before making investment decisions.
- Core data sources (Macrotrends, Finviz, etc.) are overseas websites; mainland-China reachability is unverified and may impair usability.
- Scoring thresholds (e.g., FCF/net income > 80%) have no stated methodology, and moat assessment is highly subjective; ratings reflect a single framework's perspective and are not investment advice.
- The publisher is not verified by FollowSkills; identity is unknown. This is a static source review with no executed reproduction tests.
What does this skill do, and when should you use it?
This is one standalone skill inside the Day1Global-Skills repository, located at us-value-investing/SKILL.md. It operationalizes Buffett-style value investing into a repeatable agent workflow: assess a company across ROE sustainability, debt safety, free cash flow quality, and economic moat, score each dimension 0-3 for a 12-point total, then map the score to an A/B/C/D investment rating and a structured report. Data comes from the agent's web search capability, with priority given to 10-K/10-Q filings and authoritative platforms like Macrotrends. It ships adjusted criteria for banks, REITs, and utilities, and explicitly states it does not fit pure growth or cyclical stocks.
After confirming the target company and ticker, it uses web_search to pull the current stock price, three years of ROE, debt ratios, cash flow statements, and moat-related analysis; scores four dimensions at 0-3 points each: sustained ROE above 15%/20% for 3+ years earns more, a debt-to-asset ratio under 30% earns full marks, FCF above 80% of net income signals high earnings quality, and the moat check covers brand, network effects, cost advantage, and switching costs (requiring 2+ strong moats for a full score); sums the 12-point total into ratings A (10-12), B (7-9), C (4-6), D (0-3); generates a templated report with per-dimension detail, rating, investment recommendation, and disclaimer, with data source links attached.
- A long-term individual investor wants a fundamentals check on a US stock before buying or continuing to hold it
- A casual user asks 'What do you think of stock XX?' and wants structured analysis instead of a vague opinion
- Someone who can't read financial reports wants the meaning of a company's ROE, debt, and cash flow numbers explained
- A value investor screening or comparing several companies to see who has the widest moat and who fails value criteria
- A user who already ran a deep earnings analysis wants to cross-validate with the 4-dimension value score (a synergy documented in the README)
What are this skill's strengths and limitations?
- Fully transparent scoring — thresholds and common pitfalls (leverage-inflated ROE, buybacks shrinking equity, cyclical swings) are spelled out in the skill, making results auditable
- Built-in adjusted criteria for high-leverage sectors: banks use Tier 1 Capital Ratio, REITs/utilities at 50-60% debt are treated as normal
- Requires authoritative sources (10-K/10-Q, Macrotrends, etc.) and mandates source links in the report, reducing fabrication risk
- Honest framing — explicitly notes a D rating doesn't mean the stock won't rise, positioning it as a long-term-holding framework rather than a universal picker
- Moat assessment is inherently subjective; different agents or runs may reach different conclusions
- Fixed thresholds (e.g., ROE 15%/20%) ignore systematic differences across industries
- Entirely dependent on web_search; analysis quality is unprotected when search fails or coverage is thin
- No valuation modeling, no growth/cyclical analysis, and no test suite verifying output stability
- Every report forcibly appends the authors' promotional footer, which needs handling in formal contexts
How do you install this skill?
Method 1 (auto-discovery): run npx skills add https://github.com/star23/Day1Global-Skills --all; all 5 skills install to .agents/skills/. Method 2 (Claude Code slash commands): git clone https://github.com/star23/Day1Global-Skills.git, then cp -r Day1Global-Skills/us-value-investing your-project/.claude/skills/ (per-project) or to ~/.claude/skills/ (global); restart Claude Code and invoke via /us-value-investing. The repo documents no separate dependency installation for this skill.
How do you use this skill?
No configuration after install — just ask in natural language, e.g. 'Is AAPL worth holding long-term from a value investing perspective?' or 'Help me analyze COST's fundamentals.' The skill searches for latest financial data, scores each dimension, and outputs the full report; the report language follows your input language (English and Chinese footer templates are preset). Note: the environment must have web search capability, otherwise it cannot fetch data.
How does this skill compare with similar options?
The sibling skill tech-earnings-deepdive in the same repo offers a 16-module deep dive suited to quarter-by-quarter tracking; this skill is lighter and focused on long-term value judgment, and the two can cross-validate each other. For sentiment or liquidity context, us-market-sentiment and macro-liquidity in the same repo are complementary.