@v587d/capital-generation 2.7.2 → 2.7.4

package/selected-skills/trading-skills/skills/macro-event-analysis/SKILL.md

@@ -0,0 +1,122 @@
+---
+name: macro-event-analysis
+description: Prepare for upcoming macro catalysts by identifying the events that matter, mapping the likely transmission channels, and surfacing timing risk for the user's markets or positions.
+---
+
+# Macro Event Analysis
+
+Use this skill when the user needs a practical read on upcoming macro event risk before holding positions, increasing size, or planning entries around catalysts.
+
+This skill will not:
+
+- predict the exact price reaction to a release
+- treat every calendar entry as decision-relevant
+- substitute macro theater for position-specific planning
+
+## Role
+
+Act like a macro risk analyst preparing a trader for event risk. Focus on timing, transmission channels, and scenario awareness, not on predicting the exact market reaction.
+
+## When to use it
+
+Use it when the user wants to:
+
+- understand the next 24 hours to two weeks of macro event risk
+- filter a long calendar into the few releases that really matter
+- map event risk to rates, FX, index futures, commodities, or sector leadership
+- decide whether holding through a catalyst is sensible or whether the calendar argues for caution
+
+## Inputs and context
+
+Ask for:
+
+- the market or exposure that matters: US equities, rates, FX, energy, Europe, and so on
+- the time window: today, next 48 hours, next week
+- any specific countries, central banks, or releases the user cares about
+- whether the user is planning entries, holding existing risk, or reviewing a macro-heavy week
+
+Helpful but optional:
+
+- current positioning context
+- specific holdings or watchlist names
+- events the user already knows are important
+
+Use the user's materials first: pasted event lists, screenshots, platform notes, transcripts, watchlists, or provider details already mentioned in the conversation.
+
+## If critical data is missing
+
+If the user already gave enough event information to analyze, do not fetch anything.
+
+If the timing or event slate is still incomplete:
+
+- check whether the conversation already points to a supported provider or includes usable access details
+- if the user already indicated `FMP` or `TradingEconomics`, use [references/providers/fmp.md](references/providers/fmp.md) or [references/providers/tradingeconomics.md](references/providers/tradingeconomics.md) directly
+- otherwise consult [references/data-providers.md](references/data-providers.md) and ask which supported provider they want to use
+- once the missing event details are gathered, continue the analysis and disclose the source used
+
+## Analysis process
+
+1. Build the relevant event slate for the requested window.
+2. Rank the events by likely decision impact, not by calendar length.
+3. Explain why each event matters through likely transmission channels such as yields, FX, growth expectations, inflation expectations, or risk appetite.
+4. Flag event clustering, overnight timing, and other conditions that compress decision time.
+5. Separate high-visibility events from lower-signal filler.
+6. State where the available event information is incomplete, stale, or missing consensus context.
+7. Translate the event slate into practical preparation questions: what should the user avoid assuming, monitor closely, or plan around?
+
+Use [references/interpretation-guide.md](references/interpretation-guide.md) when you need a reminder to emphasize event risk over prediction.
+
+## Core Assessment Framework
+
+Rank each event against four anchors before calling it important:
+
+- `Market Sensitivity`: does the user's market historically react to this release or central-bank signal. Example: CPI matters more for rates and duration-sensitive equities than second-tier housing data.
+- `Surprise Capacity`: is there meaningful room between prior, consensus, and current positioning. Example: payrolls with wide estimate dispersion has more surprise capacity than a low-variance calendar filler release.
+- `Transmission Speed`: how quickly the event can move yields, FX, commodities, or index futures. Example: a policy rate decision has faster transmission than a backward-looking inventory series.
+- `Timing Pressure`: does the release compress the user's decision window through overnight timing, clustering, or proximity to existing risk.
+
+Use the anchors to sort events into:
+
+- `primary`: directly relevant, high transmission, and able to change decision quality now
+- `secondary`: worth monitoring, but less likely to dominate positioning
+- `background`: context only unless the user's book is unusually sensitive
+
+## Evidence That Would Invalidate This Analysis
+
+- the event timing changes materially or the release is revised, delayed, or cancelled
+- the consensus or prior values that anchored the importance ranking turn out to be stale or missing
+- another event overtakes the stated catalyst by moving yields, FX, or risk appetite more decisively
+- the user's actual exposure changes, making the current transmission map less relevant
+- market pricing has already absorbed the catalyst so completely that the expected sensitivity is no longer plausible
+
+## Output structure
+
+Prefer this output order:
+
+1. `Event Slate`
+2. `Core Assessment Framework`
+3. `Transmission Channels`
+4. `Preparation Brief`
+5. `Evidence That Would Invalidate This Analysis`
+6. `Source And Caveats`
+
+Always return:
+
+- a prioritized event slate for the requested window
+- why each event matters for the user's market or exposure
+- the main transmission channels to watch
+- timing notes, including clustered or overnight risk
+- a short preparation brief: what to monitor, what to avoid assuming, and where caution is warranted
+- explicit source, freshness, and caveats when provider-based data is used
+
+## Best practices
+
+- do not claim that a release guarantees direction
+- do not bury missing consensus or stale timestamps
+- distinguish event importance from certainty
+- keep the output tied to the user's actual markets and holding period
+
+## Usage examples
+
+- "Use `macro-event-analysis` for the next 48 hours. I care about USD rates, SPX futures, and anything that could force shorter holding periods."
+- "Use `macro-event-analysis` for next week with a focus on US and euro area events, and tell me where the available calendar data looks thin."

package/selected-skills/trading-skills/skills/macro-event-analysis/references/data-providers.md

@@ -0,0 +1,23 @@
+# Data Providers
+
+Use this only if the user did not already provide enough event information.
+
+Supported providers:
+
+- `FMP`: economic release calendar with timing and basic consensus fields through `/stable/economic-calendar`
+- `TradingEconomics`: macro event schedule and release context through `/calendar/...`
+
+If the user already named one of these providers or already shared usable access details, use that provider path directly.
+
+Otherwise ask: "Which provider do you want to use for the missing macro event data: FMP or TradingEconomics?"
+
+Fallback strategy:
+
+- start with `FMP` when the user needs a quick event slate by date window and country
+- use `TradingEconomics` when the user needs indicator-specific history, country filtering, or source-heavy calendar context
+- if one provider lacks forecast, prior, or timing detail for a critical event, say so and ask whether to supplement with the other supported provider
+
+Provider notes:
+
+- [FMP](providers/fmp.md)
+- [TradingEconomics](providers/tradingeconomics.md)

package/selected-skills/trading-skills/skills/macro-event-analysis/references/interpretation-guide.md

@@ -0,0 +1,22 @@
+# Interpretation Guide
+
+Focus on event risk, not prediction.
+
+## What to highlight
+
+- high-impact releases
+- clustered events that compress decision time
+- events likely to affect rates, FX, index futures, or sector leadership
+- missing consensus or previous values that reduce context quality
+
+## What not to claim
+
+- that an event will move markets in a specific direction
+- that the provider has full global coverage
+- that stale timestamps are harmless
+
+## Reporting guidance
+
+- Put source and freshness near the top.
+- Distinguish scheduled events from released events.
+- If timestamps or importance fields look incomplete, say so explicitly.

package/selected-skills/trading-skills/skills/macro-event-analysis/references/providers/fmp.md

@@ -0,0 +1,28 @@
+# FMP
+
+Use this path only when the user already indicated `FMP` or chose it from the supported provider list.
+
+Use the official economic calendar endpoint for the specific window you need:
+
+```bash
+curl "https://financialmodelingprep.com/stable/economic-calendar?from=2026-03-06&to=2026-03-13&country=US&apikey=<FMP_API_KEY>"
+```
+
+Pull only the fields you need for the analysis:
+
+- `date`
+- `country`
+- `event`
+- `impact`
+- `previous`
+- `estimate`
+- `actual`
+- `lastUpdated`
+
+Filter to the user’s actual window and countries before writing the analysis.
+
+After collecting the missing facts, return to the macro analysis. Disclose that `FMP` supplied the event schedule.
+
+Official docs:
+
+- [FMP economic calendar API](https://site.financialmodelingprep.com/developer/docs/stable/economics-calendar)

package/selected-skills/trading-skills/skills/macro-event-analysis/references/providers/tradingeconomics.md

@@ -0,0 +1,40 @@
+# TradingEconomics
+
+Use this path only when the user already indicated `TradingEconomics` or chose it from the supported provider list.
+
+Use the smallest calendar request that fills the missing event facts.
+
+## Best endpoint for a country-level upcoming slate
+
+```bash
+curl "https://api.tradingeconomics.com/calendar/country/united%20states?c=<TE_API_KEY>&f=json"
+```
+
+Use this when the user wants the next relevant releases for one country and you do not need a narrow indicator filter yet.
+
+## Best endpoint for a specific indicator and date range
+
+```bash
+curl "https://api.tradingeconomics.com/calendar/country/united%20states/indicator/nonfarm%20payrolls/2026-03-01/2026-03-31?c=<TE_API_KEY>&f=json"
+```
+
+Use this when the user already identified the key release and you only need the dated event history or forward window for that indicator.
+
+Pull only the fields you need:
+
+- `Date`
+- `Country`
+- `Event`
+- `Actual`
+- `Previous`
+- `Forecast`
+- `Importance`
+- `LastUpdate`
+- `Source`
+
+Keep the result analytical. Do not mirror the whole calendar back to the user unless they asked for that explicitly.
+
+Official docs:
+
+- [TradingEconomics economic calendar by country docs](https://docs.tradingeconomics.com/economic_calendar/country/)
+- [TradingEconomics economic calendar point-in-time docs](https://docs.tradingeconomics.com/economic_calendar/point-in-time/)

package/selected-skills/trading-skills/skills/market-regime-analysis/SKILL.md

@@ -0,0 +1,117 @@
+---
+name: market-regime-analysis
+description: Analyze current market context through trend, volatility, breadth, and event backdrop so the user can choose tactics that fit the environment without relying on black-box regime claims.
+---
+
+# Market Regime Analysis
+
+Use this skill when you need a disciplined read on market context before selecting tactics, sizing, or holding period.
+
+This skill will not:
+
+- forecast market returns from a label
+- assign fake confidence percentages to limited evidence
+- replace instrument-specific trade planning or risk limits
+
+## Role
+
+Act like a market structure analyst. Your job is to classify the environment conservatively, explain the evidence, and highlight what would invalidate that view.
+
+## When to use it
+
+Use it when the user wants to:
+
+- decide whether trend-following, mean reversion, or caution is the better default
+- understand whether a market is healthy, fragile, defensive, or in transition
+- adapt size and expectations to changing breadth or volatility
+- pressure-test a market read before committing capital
+
+## Inputs and context
+
+Ask for observations about:
+
+- trend
+- volatility
+- breadth or participation
+- leadership quality or concentration
+- liquidity and event backdrop
+- relevant timeframe and market focus
+
+If the user is vague, ask for the minimum evidence needed instead of forcing a fake label.
+
+Use the user's materials first: market notes, watchlists, screenshots, chart observations, breadth notes, or any provider details already mentioned in the conversation.
+
+## If critical data is missing
+
+If the user already supplied enough evidence to classify the environment, do not fetch anything.
+
+If trend, volatility, breadth, or event context is still too thin:
+
+- check whether the user already named a supported provider or already shared usable access details
+- if they already indicated `FMP`, `TradingEconomics`, or `Polygon`, use [references/providers/fmp.md](references/providers/fmp.md), [references/providers/tradingeconomics.md](references/providers/tradingeconomics.md), or [references/providers/polygon.md](references/providers/polygon.md) directly
+- otherwise consult [references/data-providers.md](references/data-providers.md) and ask which supported provider they want to use
+- once the missing context is gathered, continue the regime analysis and disclose the source used
+
+## Analysis process
+
+1. Assess trend, volatility, breadth, and event backdrop separately.
+2. Look for agreement or conflict across those dimensions.
+3. Classify the environment conservatively.
+4. Explain which tactics fit the environment and which become fragile.
+5. State what evidence would change the classification.
+
+Use [references/regime-framework.md](references/regime-framework.md) if you need the default label set and interpretation guidance.
+
+## Core Assessment Framework
+
+Score the environment on four anchors before choosing a label:
+
+- `Trend`: higher highs and higher lows on the user's timeframe, or repeated failure at support and resistance. Example: 20-day trend up but 5-day momentum flattening means the trend anchor is positive but weakening.
+- `Volatility`: realized range and gap behavior relative to the recent norm. Example: daily ranges expanding from 1.1% to 2.0% means tactics should become more defensive even if price is still trending.
+- `Breadth`: participation across sectors, index members, or the user's watchlist. Example: index up while only megacap tech participates counts as narrow breadth, not healthy breadth.
+- `Event Backdrop`: whether macro releases, earnings clusters, or policy headlines can invalidate the read quickly. Example: CPI tomorrow and a central-bank speaker today should push the backdrop toward heavy even if tape action is calm.
+
+Use these anchors to classify:
+
+- `healthy trend`: trend positive, volatility contained or normal, breadth broad enough, event backdrop not immediately disruptive
+- `fragile trend`: trend positive, but breadth narrow or volatility elevated
+- `transition`: anchors conflict meaningfully and no single tactic deserves high conviction
+- `defensive`: trend negative or unstable, volatility elevated, breadth weak, or event backdrop heavy enough to shrink decision time
+
+## Evidence That Would Invalidate This Analysis
+
+- the next session or two reverses the trend anchor with a decisive break of the key level the analysis relied on
+- breadth expands or collapses enough to contradict the current participation read
+- volatility contracts or explodes enough to make the current tactic set inappropriate
+- a new macro or earnings event changes the backdrop from manageable to heavy, or the reverse
+- the user's timeframe changes materially, because intraday and swing regime reads should not be treated as interchangeable
+
+## Output structure
+
+Prefer this output order:
+
+1. `Regime Summary`
+2. `Core Assessment Framework`
+3. `Tactical Implications`
+4. `Evidence That Would Invalidate This Analysis`
+5. `Caveats`
+
+Always include:
+
+- regime classification in plain language
+- key evidence used
+- what weakens or invalidates the current read
+- tactical implications for risk-taking, trade duration, and expectation-setting
+
+Avoid fake certainty or theatrical confidence percentages.
+
+## Best practices
+
+- do not forecast returns from a single regime label
+- do not turn limited evidence into precise confidence scores
+- do not replace instrument-specific trade planning
+
+## Usage examples
+
+- "Use `market-regime-analysis` from these observations: uptrend intact, breadth narrowing, volatility elevated, CPI tomorrow."
+- "Use `market-regime-analysis` on my notes from this week and tell me whether trend-following or mean reversion is the safer default."

package/selected-skills/trading-skills/skills/market-regime-analysis/references/data-providers.md

@@ -0,0 +1,26 @@
+# Data Providers
+
+Use this only if the user did not already provide enough market context.
+
+Supported providers:
+
+- `FMP`: recent price action, benchmark index context, and event calendar context through quotes, light historical prices, and the economic calendar
+- `TradingEconomics`: market history for benchmark symbols and optional economic calendar checks through `/markets/historical` and `/calendar/country/...`
+- `Polygon`: strong default for recent price history and benchmark comparison when the user already uses Polygon market data
+
+If the user already named one of these providers or already shared usable access details, use that provider path directly.
+
+Otherwise ask: "Which provider do you want to use for the missing market context: FMP, TradingEconomics, or Polygon?"
+
+Fallback strategy:
+
+- start with `FMP` when you need one provider for price context plus a quick macro-event check
+- use `TradingEconomics` when cross-checking market history with macro calendar pressure
+- use `Polygon` when the user already uses Polygon and the missing context is recent price action or benchmark comparison
+- if event-backdrop context is still missing after a `Polygon` query, say so and ask whether to supplement with `FMP` or `TradingEconomics`
+
+Provider notes:
+
+- [FMP](providers/fmp.md)
+- [TradingEconomics](providers/tradingeconomics.md)
+- [Polygon](providers/polygon.md)

package/selected-skills/trading-skills/skills/market-regime-analysis/references/providers/fmp.md

@@ -0,0 +1,46 @@
+# FMP
+
+Use this path only when the user already indicated `FMP` or chose it from the supported provider list.
+
+Use it to pull only the minimum price or event context needed for the regime read.
+
+## Best endpoint for latest benchmark snapshot
+
+Use the short quote endpoint for quick current context:
+
+```bash
+curl "https://financialmodelingprep.com/stable/quote-short?symbol=SPY&apikey=<FMP_API_KEY>"
+```
+
+This is useful when you only need the latest price, change, and volume snapshot.
+
+## Best endpoint for recent trend
+
+Use the light historical endpoint for a short lookback:
+
+```bash
+curl "https://financialmodelingprep.com/stable/historical-price-eod/light?symbol=SPY&from=2026-02-20&to=2026-03-06&apikey=<FMP_API_KEY>"
+```
+
+Repeat for other symbols only if needed, such as `QQQ`, `IWM`, or sector ETFs the user already cares about.
+
+## Optional event-backdrop check
+
+If the regime read also needs macro timing context, use:
+
+```bash
+curl "https://financialmodelingprep.com/stable/economic-calendar?from=2026-03-06&to=2026-03-13&country=US&apikey=<FMP_API_KEY>"
+```
+
+## How to use the result
+
+- use quote data for the latest snapshot, not for forecasting
+- use historical prices to judge trend quality and recent volatility
+- use the calendar endpoint only if event risk is part of the missing context
+- disclose that FMP supplied part of the market context
+
+Official docs:
+
+- [FMP quote short API](https://site.financialmodelingprep.com/developer/docs/stable/quote-short)
+- [FMP historical price light API](https://site.financialmodelingprep.com/developer/docs/stable/historical-price-eod-light)
+- [FMP economic calendar API](https://site.financialmodelingprep.com/developer/docs/stable/economics-calendar)

package/selected-skills/trading-skills/skills/market-regime-analysis/references/providers/polygon.md

@@ -0,0 +1,36 @@
+# Polygon
+
+Use this path only when the user already indicated `Polygon` or chose it from the supported provider list.
+
+Polygon is strongest here for recent price action and benchmark comparison. It does not replace macro calendar context.
+
+## Best endpoint for recent benchmark history
+
+Call the official aggregate bars endpoint:
+
+```bash
+curl "https://api.polygon.io/v2/aggs/ticker/SPY/range/1/day/2026-02-20/2026-03-06?adjusted=true&sort=asc&limit=500&apiKey=<POLYGON_API_KEY>"
+```
+
+Repeat for `QQQ`, `IWM`, or sector ETFs only if the breadth or leadership question actually requires them.
+
+Pull only the fields you need:
+
+- `ticker`
+- `results[].t`
+- `results[].o`
+- `results[].h`
+- `results[].l`
+- `results[].c`
+- `results[].v`
+
+## How to use the result
+
+- use the bar series to judge trend persistence, recent range expansion, and relative benchmark strength
+- do not imply Polygon answered the macro-event question if the backdrop still depends on calendar risk
+- if event timing still matters after the price check, ask whether to supplement with `FMP` or `TradingEconomics`
+- disclose that Polygon supplied the market-price context
+
+Official docs:
+
+- [Polygon aggregate bars API](https://polygon.io/docs/rest/stocks/aggregates/custom-bars)

package/selected-skills/trading-skills/skills/market-regime-analysis/references/providers/tradingeconomics.md

@@ -0,0 +1,46 @@
+# TradingEconomics
+
+Use this path only when the user already indicated `TradingEconomics` or chose it from the supported provider list.
+
+Use it to pull only the minimum price or event context needed for the regime read.
+
+## Best endpoint for recent market trend
+
+Call the official markets historical endpoint with a short benchmark basket:
+
+```bash
+curl "https://api.tradingeconomics.com/markets/historical/spy:us,qqq:us,iwm:us?c=<TE_API_KEY>&d1=2026-02-20&d2=2026-03-06&f=json"
+```
+
+Use this when you need a quick read on trend, relative strength, or narrowing breadth across major US equity benchmarks.
+
+Pull only the fields you need:
+
+- `Symbol`
+- `Date`
+- `Open`
+- `High`
+- `Low`
+- `Close`
+
+## Optional event-backdrop check
+
+If the regime read also needs macro timing context, use the economic calendar by country endpoint:
+
+```bash
+curl "https://api.tradingeconomics.com/calendar/country/united%20states?c=<TE_API_KEY>&f=json"
+```
+
+Use only the relevant upcoming events. Do not dump the entire calendar into the output.
+
+## How to use the result
+
+- use the market history to judge trend persistence, volatility, and leadership concentration
+- use the calendar only to confirm whether event risk is quiet, normal, or heavy
+- if breadth, liquidity, or cross-asset context is still incomplete, say so directly
+- disclose that TradingEconomics supplied part of the market context
+
+Official docs:
+
+- [TradingEconomics markets historical docs](https://docs.tradingeconomics.com/markets/historical/)
+- [TradingEconomics economic calendar by country docs](https://docs.tradingeconomics.com/economic_calendar/country/)

package/selected-skills/trading-skills/skills/market-regime-analysis/references/regime-framework.md

@@ -0,0 +1,23 @@
+# Regime Framework
+
+The v1 framework uses four dimensions:
+
+- trend
+- volatility
+- breadth
+- event backdrop
+
+## Example labels
+
+- trending-risk-on
+- trending-but-fragile
+- range-bound
+- defensive-risk-off
+- transition
+
+## Interpretation guidance
+
+- Strong trend with healthy breadth and contained volatility supports trend-following tactics.
+- Strong trend with narrow leadership or unstable volatility is weaker than it looks.
+- Range conditions favor faster profit-taking and smaller expectations.
+- Heavy event risk reduces confidence in any classification.

package/selected-skills/trading-skills/skills/portfolio-concentration/SKILL.md

@@ -0,0 +1,119 @@
+---
+name: portfolio-concentration
+description: Evaluate whether a portfolio, account, or planned position is too dependent on a small number of issuers, sectors, themes, or correlated exposures before the user adds or holds more risk.
+---
+
+# Portfolio Concentration
+
+Use this skill before adding risk when the user needs to know whether a new or existing position would make the portfolio too dependent on a small set of holdings or shared drivers.
+
+This skill will not:
+
+- build a full strategic asset-allocation plan
+- decide tax strategy for concentrated holdings
+- replace entry, stop, target, or sizing decisions
+
+## Role
+
+Act like a conservative portfolio risk reviewer. Your job is to identify where diversification is weaker than it looks and where one idea may already dominate the book.
+
+## When to use it
+
+Use it when the user wants to:
+
+- check whether a new trade would overconcentrate the portfolio
+- identify hidden overlap across stocks, ETFs, sectors, themes, or employer stock
+- understand whether the book is too dependent on one catalyst, factor, or industry
+- decide whether the portfolio can absorb additional risk before moving to `position-sizing`
+
+## Inputs and context
+
+Ask for:
+
+- current holdings with approximate weights, market values, or percentages
+- whether the snapshot is the full portfolio, one account, or only the risk sleeve being reviewed
+- any planned new position or add-on size
+- whether any holding is employer stock, a narrow ETF, a thematic basket, or an illiquid position
+- the user's objective and timeframe: tactical trading book, swing portfolio, retirement account, long-term core, and so on
+
+Helpful but optional:
+
+- sector, theme, or factor tags the user already tracks
+- tax sensitivity if trimming a winner would be costly
+- whether the user already knows certain holdings overlap
+
+Use the user's materials first.
+
+If the holdings snapshot is partial, say so clearly and keep the result provisional. Do not pretend a partial account view is the whole portfolio.
+
+Work from the snapshot the user gave you. This plugin cannot read a brokerage account — the only positions it knows are the ones the user marked as held, at weights they reported themselves — so do not go and pull balances, quotes or holdings yourself: hand the exact brief (ticker, the field needed, the as-of date) to the host's own retrieval path, in a role-split host the agent that owns market data, and keep the result provisional wherever the snapshot stays partial.
+
+## Analysis process
+
+1. Reconstruct the exposure set from the user's holdings and planned change.
+2. Measure direct concentration by issuer and by top positions.
+3. Check indirect concentration from sector, theme, benchmark overlap, employer stock, or correlated assets.
+4. Flag holdings that may be harder to reduce quickly because of liquidity, tax friction, restrictions, or emotional attachment.
+5. Judge whether the proposed add would materially worsen the concentration profile.
+6. Separate concentration risk from thesis quality. A good idea can still be too large in the context of the book.
+7. End with the portfolio guardrail implication: proceed, reduce size, diversify first, or review more complete holdings before acting.
+
+Use [references/concentration-framework.md](references/concentration-framework.md) when you need the default heuristic thresholds and overlap checklist.
+
+## Core Assessment Framework
+
+Assess the portfolio on five anchors before calling it well balanced:
+
+- `Single-Issuer Exposure`: how much of the portfolio depends on one stock or bond issuer. Example: a single stock above roughly 5% deserves review; above roughly 10% is often already an elevated concentration case for many investors.
+- `Correlated Cluster Exposure`: how much of the book depends on the same sector, theme, geography, factor, or benchmark leaders. Example: owning NVDA, SMH, and a cloud or AI theme sleeve may be more concentrated than the ticker count suggests.
+- `Employment Or Identity Link`: whether the same company or theme affects both portfolio value and outside life circumstances. Example: employer stock can create a double hit to income and net worth.
+- `Exit Flexibility`: how easily the user could reduce the exposure if conditions change. Example: illiquid securities, narrow products, or restricted stock create worse concentration risk than a liquid broad ETF at the same weight.
+- `Snapshot Completeness`: whether the analysis saw enough of the user's holdings to judge true diversification. Example: a single brokerage account may understate bond exposure, international exposure, or offsetting holdings elsewhere.
+
+Use the anchors to classify:
+
+- `acceptable`: no position or correlated cluster appears dominant relative to the stated objective, and the planned add does not materially distort the book
+- `elevated`: one or more positions or clusters deserve active monitoring or a smaller add because the portfolio is becoming too dependent on a limited set of drivers
+- `too concentrated`: issuer, cluster, employer-stock, or liquidity risk is strong enough that adding exposure or holding as-is deserves a rethink before trade construction proceeds
+
+## Evidence That Would Invalidate This Analysis
+
+- the holdings snapshot was incomplete or stale enough to hide meaningful offsetting or overlapping exposures
+- the planned add size changes materially
+- a holding's sector, theme, or benchmark role was misclassified
+- tax, lock-up, or liquidity constraints were stronger or weaker than described
+- the user's objective changes from tactical trading capital to long-horizon core capital, or the reverse
+
+## Output structure
+
+Prefer this output order:
+
+1. `Concentration Summary`
+2. `Top Exposure Risks`
+3. `Overlap And Correlation Review`
+4. `Portfolio Guardrail`
+5. `What Changes The Verdict`
+6. `Next Skill`
+
+Always include:
+
+- whether the snapshot was full or partial
+- the main issuer or cluster concentration
+- any hidden overlap across stocks, ETFs, sectors, or themes
+- whether the planned add worsens concentration materially
+- any special risk from employer stock, narrow products, or illiquid holdings
+- whether the user should move next to `position-sizing`, reduce exposure first, or review a fuller portfolio snapshot
+
+A skill named above is an option for the user, not a step this role can take: only the skills the user switched on exist in this session, and the ones this plugin deliberately does not offer — placing orders, setting position size — are not reachable at all. Say what you would check next and let the user decide.
+
+## Best practices
+
+- do not equate more tickers with true diversification
+- do not ignore ETF or fund overlap with individual positions
+- do not treat employer stock like an ordinary holding
+- do not recommend a larger position just because the thesis is attractive
+
+## Usage examples
+
+- "Use `portfolio-concentration` on this book before I add more semis: NVDA 8%, SMH 10%, QQQ 12%, cash 20%, the rest broad ETFs."
+- "Use `portfolio-concentration` on my retirement account and tell me whether my employer stock and AI funds have made the portfolio more concentrated than it looks."