AlphaGBM Skills

Market research inside your AI workspace.

Find stock opportunities. Compare options strategies. Understand news. Break down research. Review investment decisions.

Website · 中文 · Catalogue · Demos · Access & usage

27 Skills · 5 core Skills + 22 focused Skills

Counts include method-reference packages, not just callable APIs. The catalogue preserves each package's access and release status; a preview is not a claim of verified production availability.

Start with the result you need

Install one Skill:

npx skills add AlphaGBM/skills --skill alphagbm-stock-research

Then ask your AI:

Use AlphaGBM to research NVDA. Explain supporting evidence, counterevidence and what could change the conclusion. Ask before using my research allowance.

The installer lets you choose your AI tool. Callable packages contain a self-contained Python 3.9+ runner; no separate AlphaGBM CLI installation is required.

Installation, local-source upgrades and version checks.

Choose a focused tool

Stocks: Stock Analysis · Dividend Strategy · Market Sentiment · Research Reports · Momentum Following · ETF Strategy · Grid Plan · Dollar-Cost Averaging · Smart Money Tracking.

Options: Scoring, volatility, Greeks, strategy comparisons, payoff analysis and risk methods. See the options directory for the exact eleven packages and their access status.

Commodities and digital assets: focused research methods, clearly labelled as references rather than live data integrations.

npx skills add AlphaGBM/skills --list
npx skills add AlphaGBM/skills --skill alphagbm-etf-strategy
npx skills add AlphaGBM/skills --skill alphagbm-research-reader

Market Sentiment includes VIX and fear indicators within one method; these are no longer separate cards. Smart Money Tracking summarizes disclosed records and never copies or places trades.

Your workspace, your AlphaGBM account

Create a personal key in your account and configure ALPHAGBM_API_KEY securely in your tool's environment. Never paste a key into a conversation or commit it to a repository.

• Installation is free; account-backed research shares the website's allowance and subscription rules. Installation does not unlock Alpha Agent or grant additional quota.
• Published research reads need no key; this does not grant access to the private research archive. Investment Review compares files you supply locally.
• Reference packages explain methods; they do not expose a live API. Runners check supported workflow contracts before applicable paid calls and never replace failures with demos.
• Preserve source dates, missing data and score definitions. Research is not a return guarantee or an order to trade.

Account plans · Verification boundary

Examples you can inspect

Every current package has an entry in the demo index: either an explicitly synthetic output fixture or a source-based example request. Neither is presented as a captured live paid response. Demo guide.

One catalogue, matching website cards

catalog/catalog.json defines names, categories, stable IDs and exact package paths. The website imports a commit-pinned copy and uses those paths in copied instructions. Core Skills live under skills/core/; focused tools live under their asset categories. Contributing.

Investor-inspired methods are maintained separately in AlphaGBM/investment-masters; they are not counted or duplicated here.

Upgrading from the flat directory

Retained package IDs and --skill install names are unchanged, but file paths have moved. Reinstall using the commands above; replace old skills// bookmarks with catalogue paths. Research-management cards and retired standalone packages are removed from the latest catalogue, not silently mapped to unrelated functions. Existing pinned versions remain available in Git history. The shared runner's compatibility commands are not new catalogue entries.

Development

python3 scripts/build_catalog.py
python3 scripts/build_catalog.py --check
python3 -m unittest discover -s scripts -p 'test_*.py'
npx skills add . --list

MIT licensed. See LICENSE.