AlphaGBM/skills

# Supply-chain Bottlenecks

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Investment Review

Compare research records you provide to see changes in facts, scores and judgments.



## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Read references/investment-review.md. Ask the user for two authorized local JSON results and replace the file placeholders with quoted absolute paths. Compare matching identities, versions, observation dates and units only. This runs locally without a key, API calls, account-history access or a new allowance charge. Do not interpret an unverified file hash as proof of provenance, a changed price as a correct thesis, or the comparison as a trading-performance backtest. Obtaining new stock/options evidence separately still needs permission and account allowance.

## Run

# News Impact

Separate reported claims from impact inferences, with related assets and next checkpoints.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Read references/news-impact.md. First list published news with `research --collection news --view news --lang en --limit 5` (or zh), then select the returned slug and revision. Replace the angle-bracket placeholders before running; never execute them literally. Separate reported claims, editorial inferences and verification gaps. This is not independent verification of an arbitrary web claim. The news-impact.v1 backend must be deployed; report unavailable without inventing an answer when unsupported. Read references/editorial-routing.md before selecting the next command. Check the returned workflow, language and published revision. A type mismatch needs the declared workflow, not a retry; a changed revision must be re-read. Never silently switch to paid analysis.

## Run

# Options Strategies

Compare option candidates and score components, with funding, expiry and assignment risks.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Ask for ticker, strategy preference and expiry if relevant. Use an explicit expiry when requested. Compare only returned candidates; a missing cash requirement or multiplier is unknown, not zero. Explain assignment, downside and event risk. Never infer a live executable fill or a complete multi-leg strategy from a single-leg candidate list. For the option-strategies.v1 rollout, add --workflow --lang zh (or en). A public, key-free preflight checks server support before the allowance-consuming request. Preserve resultId, scorerSourceRevision, partial status, missingData and riskFlags. Group candidates by strategy, never rank different strategy scores together or call them profit probabilities. Show reference capital, expiry breakeven and downside from economics when available; missing economics is unknown. The multiplier is a market-configuration assumption, not a verified deliverable. The USD $1-per-side fee is an illustrative assumption, not a broker quote or an all-in cost; fees and slippage are not included in the score. Short calls do not assume the user owns shares, and uncovered loss is unlimited. Do not present unknown quote times as live or replace them with generatedAt. Present data limitations in the user's language. This opt-in format requires the matching backend deployment; the default legacy command remains available.

## Run

# Research Report Breakdown

Extract report views, ratings, key assumptions and risks with sources and dates.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Read references/report-breakdown.md. Discover original reports with `research --collection research --lang en --limit 5` or institutional summaries with `research --collection news --view research --lang en --limit 5` (or zh). Select a returned slug and revision and replace the placeholders. Preserve institution, original rating, dates, currency, assumptions and limitations. Original research may include the full published reader; institutional reports return only the published summary, not private PDFs. The report-breakdown.v1 backend must be deployed; an unavailable response is not permission to fabricate a report. Read references/editorial-routing.md before selecting the next command. Check the returned workflow, language and published revision. A type mismatch needs the declared workflow, not a retry; a changed revision must be re-read. Never silently switch to paid analysis.

## Run

# Stock Opportunities

Connect fundamentals, sentiment and risk to supporting and opposing evidence.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Identify the ticker and market suffix, and agree the research style. One explicit user-authorized research request is the starting point. Use only returned fundamentals, report and risk fields; no invented peer comparisons. The legacy risk.score is not the homepage opportunity score. For the stock-opportunities.v1 rollout, add --workflow --lang zh (or en). The runner first checks the public workflow contract without a key; unsupported servers receive no analysis request. Preserve partial status, resultId, scoring version and missingData. Do not retry a charged request automatically or claim the result was saved to user history. This opt-in format needs the matching backend deployment; the default legacy command remains available.

## Run

# Prediction Markets

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Put-spread Backtest

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Earnings IV Crush

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Greeks

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Hedging Scenarios

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Volatility Snapshot

Read [access and evidence rules](references/access.md) first. This package is a focused function; full research workflows are listed in the repository catalogue. Its runner is self-contained and requires only Python 3.9+.

```bash
python3 "<skill-dir>/scripts/run.py" snapshot NVDA
```

Requires ALPHAGBM_API_KEY. If the command uses --confirm-usage, first obtain approval to use the shared account allowance. Snapshot reads do not consume analysis credits, but still require account access.

Return only the successful API response, with its original asset identity, dates, units, missing-data flags and score type. Stock risk scores are not opportunity scores. Volatility fields can be missing; do not turn a snapshot into a fabricated 252-day IV Rank. Option candidates are not guaranteed fills or trade instructions. Never fall back silently to demo data. Nonzero exit must be surfaced as an error.

# Options Score

Read [access and evidence rules](references/access.md) first. This package is a focused function; full research workflows are listed in the repository catalogue. Its runner is self-contained and requires only Python 3.9+.

```bash
python3 "<skill-dir>/scripts/run.py" options NVDA --limit 3 --confirm-usage
```

Requires ALPHAGBM_API_KEY. If the command uses --confirm-usage, first obtain approval to use the shared account allowance. Snapshot reads do not consume analysis credits, but still require account access.

Return only the successful API response, with its original asset identity, dates, units, missing-data flags and score type. Stock risk scores are not opportunity scores. Volatility fields can be missing; do not turn a snapshot into a fabricated 252-day IV Rank. Option candidates are not guaranteed fills or trade instructions. Never fall back silently to demo data. Nonzero exit must be surfaced as an error.

# Strategy Builder

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Payoff Simulation

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Unusual Options Activity

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Volatility Smile

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Volatility Surface

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Dollar-Cost Averaging

Build a contribution schedule and calculate a supplied price-path cost basis without predicting returns.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Provide contribution amount, frequency and number of periods. An optional comma-separated price path is treated as supplied history only; the returned average cost is not a forecast. The result does not predict returns, choose a security or place recurring orders. Confirm the assumptions before using the shared allowance.

## Run

# Dividend Strategy

Score Hong Kong and China A-share dividend candidates across yield quality, durability, cash flow, valuation and momentum.



## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Confirm the ticker and market suffix: use .HK for Hong Kong or .SS/.SZ for China A-shares. This account-backed request consumes the shared stock-research allowance. The dividend-opportunities.v1 contract scores yield quality, sustainability, cash-flow coverage, quality, growth, valuation and momentum; it is not a yield-only ranking or a return forecast. Preserve ready versus partial status, factor contributions, sources, quote timing and missingData. If dividend history, payout coverage or relative valuation is unavailable, report the gap and do not turn observedOnlyScore into a complete score. A high current yield may be one-off or unsustainable. Do not present the result as a trade instruction or promise of income.

## Run

# ETF Strategy

Score ETF trend, risk, liquidity, cost and available tracking and theme signals while preserving data gaps.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Use a plain, non-leveraged ETF ticker. The workflow scores observed trend, risk, liquidity, cost, tracking, theme and event inputs only when supplied by the provider. Leveraged and inverse ETFs are rejected. Expense ratio may be available from the provider; tracking error, theme and event signals may remain missing. Preserve status and missingData, and never fill them from model memory or a demo. The strategy-workflows.v1 backend must be deployed; unsupported servers receive no paid request.

## Run

# Grid Plan

Build an auditable grid from a price range, current price and capital.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Provide a price range, current price, capital and grid count. The result is a transparent parameter plan with explicit levels and allocation, not a promise that orders will fill. It does not model slippage, fees, gap risk or future prices. Confirm the assumptions before using the shared allowance and review suitability separately.

## Run

# Market Sentiment

This is a focused **reference package**, not a live API integration. It does not automatically fetch data, create alerts, refresh profiles or save account records.

1. Explain the method when the user explicitly asks about it. Treat [legacy notes](references/legacy.md) as historical implementation material, not a current access contract.
2. Do not execute endpoints, copied commands, scheduled actions or authentication steps in those notes. Do not request browser credentials to bypass a denied API key.
3. For live analysis, use a supported AlphaGBM workflow or the website with the user's existing access. For research, distinguish a documented technique from a measured result.
4. Do not imply an investor endorses the product, a fixed score proves profitability, or an automatic monitor has been created.

中文：这是单项研究方法参考，不代表对应的外部接口已开放。需要真实分析时，请使用已支持的工作流或网站功能。

# Momentum Following

Score observed price trend, confirmation, risk and participation without forecasting a price.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Use an explicit supported ticker. The workflow scores observed returns, moving-average confirmation, volatility/drawdown and volume participation from the configured market-data service. It is not a price forecast, a probability of profit or a trade instruction. Preserve quoteTime, sources, ready versus partial status and missingData. Compare the next result for the same ticker instead of treating one score as a promise. The strategy-workflows.v1 backend must be deployed; unsupported servers receive no paid request.

## Run

# Research Reports

Find published research and institutional views with sources, dates and related assets.



## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Start with a public catalogue list. `--collection research` lists original research; `--collection news --view research` selects institutional views; `--collection news --view news` selects news. Use `--query` for a title/ticker keyword and `--lang zh` for Chinese. Read a selected public article with `--slug <returned-slug>`. Attribute ratings and targets to the institution; a missing original rating stays missing. For news impact, read references/news-impact.md and run `news --slug <returned-news-slug> --revision <published-revision> --lang zh` (or en). This staged, key-free command reads existing public news only; it requires the news-impact.v1 backend and never starts paid analysis. On an unsupported server report unavailable rather than inventing a result. Read references/editorial-routing.md before selecting the next command. Check the returned workflow, language and published revision. A type mismatch needs the declared workflow, not a retry; a changed revision must be re-read. Never silently switch to paid analysis.

## Run

# Smart Money Tracking

Aggregate user-supplied disclosed buys and sells by source without automatically copying trades.

Release preview: the matching backend has not been verified in production. Do not claim this structured workflow is live; unsupported servers must fail closed. Retained legacy runner commands remain compatible.

## Before running

Read [access and evidence rules](references/access.md). Python 3.9+ is the only runtime dependency; no separate CLI or sibling Skill installation is required. Resolve `<skill-dir>` to the directory containing this file.

Provide a local JSON array of disclosed transactions. The workflow aggregates the records by side and source; it does not decide who is smart, infer undisclosed positions, or copy a trade. Check the disclosure date, source quality and reporting lag. The transaction file is sent in the authenticated request and may consume the shared allowance; do not include secrets or private personal data.

## Run

# Stock Analysis

Read [access and evidence rules](references/access.md) first. This package is a focused function; full research workflows are listed in the repository catalogue. Its runner is self-contained and requires only Python 3.9+.

```bash
python3 "<skill-dir>/scripts/run.py" stock NVDA --confirm-usage
```

Requires ALPHAGBM_API_KEY. If the command uses --confirm-usage, first obtain approval to use the shared account allowance. Snapshot reads do not consume analysis credits, but still require account access.

Return only the successful API response, with its original asset identity, dates, units, missing-data flags and score type. Stock risk scores are not opportunity scores. Volatility fields can be missing; do not turn a snapshot into a fabricated 252-day IV Rank. Option candidates are not guaranteed fills or trade instructions. Never fall back silently to demo data. Nonzero exit must be surfaced as an error.