@v587d/capital-generation 2.7.11 → 2.8.0

package/selected-skills/investment-skills/skills/buffett-investment-framework/references/09-workflows.md

@@ -0,0 +1,96 @@
+# Workflows
+
+Card sequences, required inputs, and the output contract for the three task workflows routed from
+*Choose the Workflow* in `SKILL.md`. Open the workflow you are running before applying
+cards; each part is sized to arrive complete in one `read`.
+
+## Run Financial Analysis
+
+Require exact filings and periods, policies and notes, segment schedules, cash-flow and
+capital-expenditure data, and material financing and obligation disclosures.
+Treat market price as a separate input only when valuation is in scope.
+
+1. Apply `F01` and `F05` to inventory the package and reconcile reported results into
+   comparable segments without discarding genuine costs.
+2. Apply `F02` to bridge reported earnings to a range of owner earnings.
+   Add `F03` when maintenance, growth, restricted, or distributable capital must be
+   separated. Show every maintenance-capital, working-capital, compensation, and
+   recurring-cost judgment.
+3. Apply `F06`, and `B06` when prospective reinvestment matters, to separate historical
+   returns from the future reinvestment runway.
+4. Add `F07`, `C02`, `R02`, and `R03` as needed to inventory obligations and stress
+   financing, liquidity, collateral, and nonlinear claims.
+5. Add `F04` for material partial ownership and only the triggered `S01`–`S11` overlay
+   for industry-specific accounting or risk.
+
+Return a package inventory, reported-to-analytical bridge, normalized segment base,
+owner-earnings range, capital-return calculations, obligation ledger, assumptions,
+sensitivities, and blocked calculations.
+
+## Review an Investment Memo
+
+Require the complete memo and appendices, cited primary sources and models, decision
+question, horizon, business and management claims, valuation bridge, alternatives,
+risks, and invalidation conditions.
+
+1. Apply `D01` and `D03` to test estimability, separate facts, calculations, and
+   inference, and seek disconfirming evidence.
+   Add `D02` when ownership framing is material.
+2. Apply `B02` to test a claimed moat as a causal mechanism.
+   Add `B01` and only the relevant `B03`–`B07` cards when the memo’s claims require
+   them.
+3. Add only material `M01`–`M08` and `C01`–`C09` cards; use `C09` when a transaction is
+   central.
+4. Apply `V01`, `V05`, and `V06` to reconstruct value and expose sensitivities.
+   Add `V03` when the normalized base is material and `V04` when projected growth
+   contributes materially to value; use `D04` when price action is being treated as
+   business evidence.
+5. Apply `R01` and `R07` to trace owner harm and invalidation.
+   Add `R02` or `R03` only for financing or contingent paths actually present.
+
+Return one row per material claim: claimed consequence, evidence and source quality,
+applicable cards, supported/challenged/unresolved disposition, counterevidence or
+alternate mechanism, and evidence still needed.
+
+## Evaluate a Thesis
+
+Require explicit thesis components and horizon, primary evidence for the claimed
+business and management mechanisms, valuation inputs, financing and risk evidence,
+market price when price versus value is asked, and failure conditions.
+
+1. Apply `D02`, `D03`, and `R07` to rewrite the thesis as testable components while
+   preserving the author’s meaning and invalidation conditions.
+2. Apply `B01`, `B02`, `B06`, and `B07` as relevant to test business mechanisms,
+   incremental-return economics, and structural change.
+3. Add only the `M01`–`M08` and `C01`–`C09` cards needed by stated management or
+   allocation dependencies.
+4. Apply `V01`, `V03`, `V04`, `V05`, and `V06` as relevant to reconstruct value from
+   supplied evidence and keep quality, value, and price as separate judgments.
+5. Add `R01`, `R02`, `R03`, and a triggered `S01`–`S12` overlay to trace owner-harm,
+   financing, contingent, and specialized risks.
+
+Return a component map, evidence and counterevidence table, mechanism tests, valuation
+assumptions and sensitivities, owner-harm paths, invalidation indicators, missing
+evidence, and a separate disposition for each component.
+Do not aggregate component dispositions into an investment recommendation.
+
+## Use This Output Contract
+
+Every response must contain:
+
+1. **Question and scope:** task intent, horizon, entity, and exclusions.
+2. **Evidence receipt:** supplied primary evidence, periods, missing inputs, and source
+   limitations.
+3. **Card load:** five to 12 IDs and why each was selected.
+4. **Analytical work:** sourced bridges, calculations, mechanism tests, and
+   sensitivities.
+5. **Findings:** supported, challenged, or unresolved, one decision question at a time.
+6. **Counterevidence and alternatives:** strongest contrary account and competing
+   mechanism.
+7. **Limits and blocked branches:** missing evidence and uncertainty that affect the
+   result.
+8. **Monitoring and invalidation:** observations that would strengthen, weaken, revise,
+   or retire each material finding.
+
+End with an analytical summary, not a buy, sell, hold, entry-price, position-size, or
+trade instruction.

package/selected-skills/investment-skills/skills/buffett-investment-framework/references/10-method.md

@@ -0,0 +1,12 @@
+# Method and Provenance
+
+The 65 cards are an editorial synthesis of published Buffett and Berkshire writings.
+Source material was decomposed into claims, definitions, tactics, examples, and
+references; recurring decision questions were inventoried; overlaps were reconciled by
+analytical consequence; and the result was organized into eight modules and three task
+workflows. The cards provide representative, not exhaustive, coverage of the source
+material. Each card includes one to three readable source notes, abbreviated
+corroboration citations reconciled against the wider published record and resolved in
+the [source key](references/00-source-key.md), and its own preserved limits.
+The framework is not affiliated with or endorsed by Warren Buffett or Berkshire
+Hathaway.

package/selected-skills/investment-skills/skills/buffett-investment-framework/references/11-deterministic-router.md

@@ -0,0 +1,11 @@
+# Deterministic Router
+
+Use the bundled deterministic router when the host gives this role a shell **and** this skill's directory is reachable. Where either is missing — a role-split host that grants no shell, or an install without `scripts/` — skip the router and open the `references/` file for the chosen module directly, using the mapping in *Choose the Workflow* in `SKILL.md`. The router only decides which file to read; it holds no analysis of its own, so nothing is lost by reading the file.
+From the skill directory, run:
+
+```bash
+python3 scripts/framework.py route --intent financial-analysis --topic insurance
+python3 scripts/framework.py route --intent memo-review --topic acquisition
+python3 scripts/framework.py show --id C08
+python3 scripts/framework.py validate
+```

package/selected-skills/peter-lynch-skill/UPSTREAM.md

@@ -60,6 +60,7 @@
   它不跨 Agent、不改本插件的角色分流，只在本颗的两条工作流之间选择（`china-stock-research-orchestrator`
   与 InvestSkill 的 `full-report` 是另一类东西）。也不触发 §7（报告交付链路已删除）与 §6.1（出图只走
   `render_chart`）：这颗的输出是**最终答复里的 markdown 报告**，不落文件、不出图。
+- Visibility layering（2026-10-14，§8.6 第三个改写理由）：这颗本来就有两层厚正文。第一层是 `SKILL.md`（8,777 字符）——流程六步、research brief 模板与失败处置搬进 `references/10-workflow.md`，入口只留 Routing 与 Required inputs。第二层是它上游自交的指南：`references/deep/methodology.md`（18,230）、`deep/web-research.md`（16,426）、`quick/web-research.md`（11,670）、`quick/methodology.md`（7,598）都超一次 `read` 的可见上限，各按 `##` 边界切成 2–4 册，**原路径保留成一份几百字符的索引**——那些指南之间互相点名（一处 workflow 指六份指南），留索引就不必改十几处指针，旧引用照用，模型多开一次几百字符的 read 而不是又开一本厚册。整棵树一共 23 份 `.md`，逐行核对原树**零行丢失**（入口那句 "The source tiers below" 改成点名 `references/10-workflow.md` 是唯一改写的一行），总量 78,244 → 85,955 字符（只增不减），每份都按 `read` 渲染后长度 ≤ 8,192 核过（判据与建模在 `test/selected-skills-content.test.mjs`）。MIT，不需要逐文件许可声明行。
 - Maintenance snapshot (checked 2026-10-11): 全仓两次提交，最后一次 2026-06-28，之后 105 天无提交；
   单人仓库、无 issue 响应记录。因此标 `watchlist`——与 `china-stock-research-skills` 同一口径
   （内容可用、维护薄，启用前请使用者自行复核来源），而不是内容缺陷。

package/selected-skills/peter-lynch-skill/skills/peter-lynch-skill/SKILL.md

@@ -47,96 +47,22 @@ Determine the workflow from the user's message:
 
 If neither a company nor ticker can be inferred from the request, ask one concise clarification question.
 
-The source tiers below (aggregator profile pages, SEC / regulator filings, annual reports) are US and overseas-listing conventions. In an A-share session the same analysis must be asked for in A-share terms instead — disclosure platforms, no Form 4 / 13F, no wash-sale rules — and that is decided by the host's retrieval channel, not by this skill.
+The source tiers in `references/10-workflow.md` (aggregator profile pages, SEC / regulator filings, annual reports) are US and overseas-listing conventions. In an A-share session the same analysis must be asked for in A-share terms instead — disclosure platforms, no Form 4 / 13F, no wash-sale rules — and that is decided by the host's retrieval channel, not by this skill.
 
 Do **not** ask the user for an annual report PDF or API keys.
 
-## Workflow overview
+## Read the Parts
 
-Copy this checklist and track progress:
+Everything below the surface of this file lives in `references/`. Open a part with `read`,
+resolved against this skill's base directory, and open it **whole — with no line window**:
+reporting a rule or a template out of a slice of a part is how a file gets described as read
+when it was only skimmed. Each part is sized to arrive complete in one `read`. If a receipt
+shows `[... tool result middle pruned ...]`, re-read it with `offset` before concluding
+anything is missing.
 
-```
-Task Progress:
-- [ ] Step 1: Route (quick or deep)
-- [ ] Step 2: Read workflow-specific web research guide
-- [ ] Step 3: Gather data and compile structured research brief
-- [ ] Step 4: Read workflow prompts
-- [ ] Step 5: Write report from the brief only
-- [ ] Step 6: Append disclaimer
-```
+- `references/10-workflow.md` — Workflow overview; Research brief; Failure handling
 
-### Step 1: Route
+Loading this file is not reading the skill: the steps, templates, and thresholds exist only
+in the parts above. Say which parts you opened, name the ones you did not, and never report a
+step or a template you have not read in this session.
 
-Select **Quick** or **Deep** using the routing table above.
-
-### Step 2: Read workflow-specific web research guide
-
-Read only the workflow-specific web guide:
-
-- **Quick:** [references/quick/web-research.md](references/quick/web-research.md)
-- **Deep:** [references/deep/web-research.md](references/deep/web-research.md)
-
-Read [references/filing-rules.md](references/filing-rules.md) only when regulator/XBRL data, ADRs, secondary listings, specialized financial companies, or accounting-basis issues matter. Do not read it for normal Quick Research unless the Quick web guide specifically triggers it.
-
-### Step 3: Gather data and compile research brief
-
-Asking the host's own retrieval path for the data (in a host that splits roles: delegate it with the brief, do not browse from here):
-
-1. Follow the **workflow-specific** source plan, retrieval brief, stopping rule, and research brief template in `references/quick/web-research.md` or `references/deep/web-research.md`.
-2. Respect the workflow-specific default budgets unless an override in that workflow guide is triggered — the budget is how many sources you ask the channel for, not how many you go and open yourself:
-   - **Quick:** practical default of 1-3 reputable aggregator sources, 2-3 retrieval queries, and 2-3 page fetches; expand only when another aggregator is needed to complete the existing Quick methodology.
-   - **Deep:** 5-7 sources, 6-8 retrieval queries, 7-10 page/document fetches, with the default source set kept as small as the report can support.
-3. For **Quick**: gather the minimum viable evidence needed for a responsible SKIP / WATCHLIST / DEEP RESEARCH decision, then stop.
-4. For **Deep**: follow the annual report discovery procedure in `references/deep/web-research.md` — ask for the latest full annual report on the company's investor relations site (HTML or PDF); regulator filings are the fallback.
-5. Do not ask for a separate retrieval per fiscal year for normal historical metrics. For Quick, use aggregator tables. For Deep, use the annual report, 10-K / 20-F, regulator filing, or regulator facts feed as the historical base.
-6. Do not print research narration, retrieval play-by-play, or the full research brief unless the user asks for it.
-7. Capture period, currency, units, date gathered, and source ID(s) for every important number and claim. Each source ID must resolve to a URL in the Source Ledger. Filing labels may be included, but they do not replace a ledger URL.
-8. Do **not** invent financial data.
-
-### Step 4: Read workflow prompts
-
-**Quick path** — read both before writing:
-
-- [references/quick/system-prompt.md](references/quick/system-prompt.md)
-- [references/quick/methodology.md](references/quick/methodology.md)
-
-**Deep path** — read both before writing:
-
-- [references/deep/system-prompt.md](references/deep/system-prompt.md)
-- [references/deep/methodology.md](references/deep/methodology.md)
-
-### Step 5: Write report
-
-Use the research brief as the only factual context for the report. Do not include the full research brief in the final report unless the user asks for it.
-
-Every final report must cite sources with links:
-
-- Any company description, financial number, valuation metric, market data point, management claim, industry claim, competitor claim, ownership claim, or news/update claim needs a source.
-- Use source lines directly below the relevant bullet or paragraph: `source: [Source name](https://...)`.
-- Multiple sources may be separated with semicolons.
-- Do not cite a source that is not in the research brief Source Ledger.
-- Expand any internal source IDs from the brief into full linked `source:` lines in the final report.
-- Source lines do not count toward Quick or Deep word limits.
-
-Internally prefix the analysis context with:
-
-```
-Data gathered: YYYY-MM-DD
-
-## Research brief
-
-{full research brief markdown}
-```
-
-### Step 6: Append disclaimer
-
-Read [references/disclaimer.md](references/disclaimer.md) only when the report body is ready. After the report body, add a horizontal rule and append the full contents of the disclaimer.
-
-## Failure handling
-
-- If the host's retrieval path returns nothing for a field you need: name the missing fields and stop the analysis that depends on them — do not fabricate data.
-- If a ticker cannot be resolved: inform the user clearly.
-- If the annual report cannot be found after all fallbacks in `references/deep/web-research.md`: note the gap in the research brief and proceed using other verified sources — do **not** ask the user for a PDF.
-- If a specific metric is unverified: say you do not know; do not estimate or invent numbers.
-- If retrieved claims lack source ID(s) that resolve to Source Ledger URL(s): do not include them in the report.
-- If live market data cannot be verified from the current source: for Quick, ask the channel for another reputable aggregator; for Deep, use the workflow's market-data fallback rules. Do not use stale or unsourced values.

package/selected-skills/peter-lynch-skill/skills/peter-lynch-skill/references/10-workflow.md

@@ -0,0 +1,94 @@
+# Workflow and Research Brief
+
+Part 1 of 1 for `peter-lynch-skill` — the six steps, the research brief template, and failure handling
+See `SKILL.md` for what to read and when; this part is procedure, not routing.
+
+## Workflow overview
+
+Copy this checklist and track progress:
+
+```
+Task Progress:
+- [ ] Step 1: Route (quick or deep)
+- [ ] Step 2: Read workflow-specific web research guide
+- [ ] Step 3: Gather data and compile structured research brief
+- [ ] Step 4: Read workflow prompts
+- [ ] Step 5: Write report from the brief only
+- [ ] Step 6: Append disclaimer
+```
+
+### Step 1: Route
+
+Select **Quick** or **Deep** using the routing table above.
+
+### Step 2: Read workflow-specific web research guide
+
+Read only the workflow-specific web guide:
+
+- **Quick:** [references/quick/web-research.md](references/quick/web-research.md)
+- **Deep:** [references/deep/web-research.md](references/deep/web-research.md)
+
+Read [references/filing-rules.md](references/filing-rules.md) only when regulator/XBRL data, ADRs, secondary listings, specialized financial companies, or accounting-basis issues matter. Do not read it for normal Quick Research unless the Quick web guide specifically triggers it.
+
+### Step 3: Gather data and compile research brief
+
+Asking the host's own retrieval path for the data (in a host that splits roles: delegate it with the brief, do not browse from here):
+
+1. Follow the **workflow-specific** source plan, retrieval brief, stopping rule, and research brief template in `references/quick/web-research.md` or `references/deep/web-research.md`.
+2. Respect the workflow-specific default budgets unless an override in that workflow guide is triggered — the budget is how many sources you ask the channel for, not how many you go and open yourself:
+   - **Quick:** practical default of 1-3 reputable aggregator sources, 2-3 retrieval queries, and 2-3 page fetches; expand only when another aggregator is needed to complete the existing Quick methodology.
+   - **Deep:** 5-7 sources, 6-8 retrieval queries, 7-10 page/document fetches, with the default source set kept as small as the report can support.
+3. For **Quick**: gather the minimum viable evidence needed for a responsible SKIP / WATCHLIST / DEEP RESEARCH decision, then stop.
+4. For **Deep**: follow the annual report discovery procedure in `references/deep/web-research.md` — ask for the latest full annual report on the company's investor relations site (HTML or PDF); regulator filings are the fallback.
+5. Do not ask for a separate retrieval per fiscal year for normal historical metrics. For Quick, use aggregator tables. For Deep, use the annual report, 10-K / 20-F, regulator filing, or regulator facts feed as the historical base.
+6. Do not print research narration, retrieval play-by-play, or the full research brief unless the user asks for it.
+7. Capture period, currency, units, date gathered, and source ID(s) for every important number and claim. Each source ID must resolve to a URL in the Source Ledger. Filing labels may be included, but they do not replace a ledger URL.
+8. Do **not** invent financial data.
+
+### Step 4: Read workflow prompts
+
+**Quick path** — read both before writing:
+
+- [references/quick/system-prompt.md](references/quick/system-prompt.md)
+- [references/quick/methodology.md](references/quick/methodology.md)
+
+**Deep path** — read both before writing:
+
+- [references/deep/system-prompt.md](references/deep/system-prompt.md)
+- [references/deep/methodology.md](references/deep/methodology.md)
+
+### Step 5: Write report
+
+Use the research brief as the only factual context for the report. Do not include the full research brief in the final report unless the user asks for it.
+
+Every final report must cite sources with links:
+
+- Any company description, financial number, valuation metric, market data point, management claim, industry claim, competitor claim, ownership claim, or news/update claim needs a source.
+- Use source lines directly below the relevant bullet or paragraph: `source: [Source name](https://...)`.
+- Multiple sources may be separated with semicolons.
+- Do not cite a source that is not in the research brief Source Ledger.
+- Expand any internal source IDs from the brief into full linked `source:` lines in the final report.
+- Source lines do not count toward Quick or Deep word limits.
+
+Internally prefix the analysis context with:
+
+```
+Data gathered: YYYY-MM-DD
+
+## Research brief
+
+{full research brief markdown}
+```
+
+### Step 6: Append disclaimer
+
+Read [references/disclaimer.md](references/disclaimer.md) only when the report body is ready. After the report body, add a horizontal rule and append the full contents of the disclaimer.
+
+## Failure handling
+
+- If the host's retrieval path returns nothing for a field you need: name the missing fields and stop the analysis that depends on them — do not fabricate data.
+- If a ticker cannot be resolved: inform the user clearly.
+- If the annual report cannot be found after all fallbacks in `references/deep/web-research.md`: note the gap in the research brief and proceed using other verified sources — do **not** ask the user for a PDF.
+- If a specific metric is unverified: say you do not know; do not estimate or invent numbers.
+- If retrieved claims lack source ID(s) that resolve to Source Ledger URL(s): do not include them in the report.
+- If live market data cannot be verified from the current source: for Quick, ask the channel for another reputable aggregator; for Deep, use the workflow's market-data fallback rules. Do not use stale or unsourced values.

package/selected-skills/peter-lynch-skill/skills/peter-lynch-skill/references/deep/methodology-1.md

@@ -0,0 +1,220 @@
+## 1) Business Model and Understandability
+
+**Output format: maximum 400 words**
+
+### 1.1 Business Model in Plain Language
+
+Answer:
+- What does the company actually sell?
+- Who are the main customers?
+- How does the company make money?
+- In which geographies does the company operate?
+
+Write this so a 12-year-old could understand it.
+
+### 1.2 Business Structure Analysis
+
+- First, analyze earnings (profit) structure by segment and product if data is available.
+
+- If segment-level profit data is available:
+    - show % contribution to total operating income or net income
+
+- If not available:
+    - fall back to revenue structure (with %)
+
+- Highlight:
+    - the most profitable segment and product
+    - whether profits are concentrated or diversified
+
+### 1.3 Hype Check
+
+- Is the business thesis based on real business performance or just hype?
+- If the exciting product/division is **<10% of total revenue**, question whether it can meaningfully drive EPS.
+
+**Warning:**  
+If the thesis relies on a revolutionary technology, miracle drug, or “next big thing” narrative without meaningful current revenue contribution, apply extreme skepticism.
+
+### 1.4 Earnings Drivers: Controllable vs External
+
+Identify **1–4 key current earnings drivers** and the drivers expected to shape future earnings.
+
+For each driver:
+- briefly describe it
+- label it as:
+  - **Mostly controllable** — mainly influenced by management decisions
+  - **Mostly external** — mainly driven by market, macro, commodity, regulation, or industry conditions
+
+Guideline: stability and continuity of earnings drivers are preferred because they improve predictability.
+
+### 1.5 Business Understandability Check
+
+Classify business complexity:
+
+- **Easy** — simple business model; revenue drivers and risks are clear
+- **Moderate** — some industry-specific metrics or dynamics require extra work
+- **Hard** — complexity, regulation, accounting, or technical factors materially affect valuation
+
+If classified as **Hard**, explicitly state what must be understood before investing.
+
+### 1.6 Investment Story & Invalidation Conditions
+
+- What is the one-sentence thesis for this business?
+- What are **2–3 conditions** that would kill the thesis? Use a numeric list.
+- How will you track whether the story is unfolding?
+
+## 2) Diagnostic Profile (No Single Category)
+
+**Output format: maximum 250 words**
+
+Report the following traits independently.  
+For each trait, give:
+- **State**
+- **Conviction**: Low / Medium / High
+- **Evidence**: 1 short sentence
+
+Use this compact bullet format:
+
+- Cyclicality: [state] | Conviction: [Low/Medium/High] | Evidence: ...
+- Turnaround status: [state] | Conviction: [Low/Medium/High] | Evidence: ...
+- Growth profile: [state] | Conviction: [Low/Medium/High] | Evidence: ...
+- Business maturity: [state] | Conviction: [Low/Medium/High] | Evidence: ...
+- Asset angle: [state] | Conviction: [Low/Medium/High] | Evidence: ...
+- Dominant analytical lenses: [list up to 2 traits that should dominate the rest of the analysis]
+
+### 2.1 Cyclicality
+
+States:
+- **High**
+- **Moderate**
+- **Low**
+
+Rules:
+- **High** if:
+  - at least **2 full EPS cycles** in the last 10 years (**peak → trough → peak**),
+  - peaks had positive EPS,
+  - swings were **≥30%**
+- **Moderate** if:
+  - 1 clear cycle, or
+  - repeated earnings swings of roughly **15–30%**, or
+  - the industry is clearly cyclical but the financial history is too short for a High rating
+- **Low** otherwise
+
+### 2.2 Turnaround Status
+
+States:
+- **None**
+- **Active**
+- **Recovering**
+
+Rules:
+- **Active** if any of the following are true:
+  - EPS declined **2+ consecutive years**
+  - revenue contracted **>5%** and margins compressed
+  - EPS is **≥30% below prior peak**
+  - and the deterioration is **not mainly normal cyclicality**
+- **Recovering** if the business recently met Active criteria but revenue, margins, or EPS have now stabilized or started improving
+- **None** otherwise
+
+### 2.3 Growth Profile
+
+States:
+- **Fast**
+- **Stalwart-like**
+- **Slow**
+- **Moderate/Other**
+- **Declining**
+
+Rules:
+- **Fast** if:
+  - **2Y EPS CAGR ≥20%**
+  - revenue growth confirms
+  - growth is **not mainly a cyclical rebound**
+  - growth is **not merely recovery from a distressed base**
+- **Stalwart-like** if:
+  - **2Y EPS CAGR = 8–12%**
+  - margins are stable
+  - no 2-year EPS decline
+  - no revenue contraction >5%
+  - and cyclicality is **not High**
+- **Slow** if:
+  - **2Y EPS CAGR = 2–5%**
+  - and the business appears **large/mature**
+  - and there is no obvious distress
+- **Moderate/Other** if results fall between bands or are mixed
+- **Declining** if EPS CAGR is below 2% or negative
+
+**Guardrail:** If EPS growth is mostly from buybacks, cost cutting, or recovery off a depressed base, say growth quality is weaker.
+
+### 2.4 Business Maturity
+
+States:
+- **Early**
+- **Scaling**
+- **Mature**
+
+Rules:
+- **Early** — still proving the model; concentrated geography/product; evidence base limited
+- **Scaling** — model proven and expanding into more markets/products; meaningful whitespace remains
+- **Mature** — broad penetration, lower whitespace, and capital allocation/dividends/buybacks matter more
+
+### 2.5 Asset Angle
+
+States:
+- **None**
+- **Possible**
+- **Strong**
+
+Rules:
+- **Strong** if verifiable after-tax asset value exceeds market cap by **≥30%**
+- **Possible** if hidden assets appear meaningful but valuation or catalyst is incomplete
+- **None** otherwise
+
+### 2.6 Dominant Analytical Lenses
+
+Choose **up to 2** traits that should dominate the rest of the analysis.  
+Examples:
+- High cyclicality
+- Active turnaround
+- Recovering turnaround
+- Fast growth
+- Strong asset angle
+- Mature slow-growth economics
+
+## 3) Long-Term Track Record
+
+**Output format: maximum 220 words**
+
+### 3.1 Revenue & EPS Trend
+
+Summarize:
+- revenue direction and rough CAGR band
+- earnings per share (EPS) consistency and volatility
+- major drops or spikes
+
+### 3.2 Earnings Predictability
+
+Rate one:
+- **Rock steady** — net income grows or holds in 4–5 of 5 years; no single year drops more than ~20%; operating cash flow confirms
+- **Moderate fluctuation** — net income swings roughly ±20–50%, or net income is noisy but operating cash flow is steadier
+- **Highly volatile** — net income swings more than 50% in multiple years, or both net income and operating cash flow are erratic
+
+If volatility is mainly due to cyclicality, say so explicitly rather than treating it as a generic red flag.
+
+### 3.3 Profitability
+
+Primary metric:
+
+- **Pretax profit margin** (rising, stable, falling?)
+
+Secondary:
+
+- Return on Equity (ROE) (direction and stability)
+
+Avoid reliance on ROIC unless clearly labeled as supplemental.
+
+### 3.4 Share Dilution
+
+State whether share count is:
+- materially shrinking
+- stable
+- heavily diluting