CoTrading Skills
Pick a concrete task, see what it produces, then take the method into your own AI tool. A Skill is a reusable method package.
No public version has completed host task acceptance yet. Read the teaching samples below; each detail page reports its actual release state.
Concrete methods and synthetic teaching samples, with host acceptance still pending. Package execution and task completion in an AI host are verified separately.
Compare two reporting periods, identify material changes and return to the original evidence.
Get: Change table, source references and follow-up questions
Bring: Company identity, two comparable reports and one question. Use readable PDFs or text with source locations.
View sample & usage →Put new evidence beside your original reasons and see what still holds.
Get: Support, counterevidence, unknowns and next checkpoints
Bring: Your original dated thesis, an as-of date and new sources. No position size, cost basis or brokerage access is required.
View sample & usage →Make a trading idea explicit and find the gaps before execution or testing.
Get: Rule card, cost assumptions and data requirements
Bring: Your idea and asset scope. Missing frequency, weights, execution timing or costs become questions first.
View sample & usage →Run defined rules through a real calculation and compare benchmark, costs and drawdown.
Get: Experiment report, equity records, costs and limits
Bring: Confirmed rules and confirmation record, prices.csv, a data manifest and complete valuation calendar. Requires a Python 3.12 environment.
View sample & usage →Try “earnings” or “ETF”, or clear your filters.
Start with one sample, follow its steps, then bring your own materials. There is no need to install every method.
Start with an earnings review →Method packages are free. Your AI account, model usage and chosen data services may have separate charges.
