skills.rest
yaojingang
yaojingangCOMMUNITY
Explore all skills by this author →
Website𝕏@yaojingangGitHub

yao-kelly-skill

Generates conservative Kelly-criterion resource allocation reports as JSON and HTML.

1.3k|146|Updated Apr 6, 2026
Search Tags:#risk management#resource allocation#position sizing#html report#fractional kelly#Kelly criterion#decision report
One-click install
npx skills add https://github.com/yaojingang/yao-open-skills --skill yao-kelly-skill
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: yao-kelly-skill
Source: https://github.com/yaojingang/yao-open-skills/tree/main/skills/yao-kelly-skill
Command: npx skills add https://github.com/yaojingang/yao-open-skills --skill yao-kelly-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

It turns uncertain invest, bet, or allocate questions into a disciplined allocation plan that states whether to allocate, how much to allocate, and when to add or stop, instead of leaving users with a raw Kelly formula.

Core Features & Use Cases

  • Conservative Kelly Sizing: Computes full Kelly fractions, then applies fractional Kelly, confidence discounts, dependence haircuts, and total exposure caps to produce a safe execution ceiling.
  • Multi-Turn Intake Loop: Scores decision readiness each round, asks only 1-3 high-impact follow-up questions, and stops asking once the action class is stable.
  • JSON + HTML Reports: Runs scripts/kelly_allocation_report.py for canonical JSON output and scripts/generate_html_report.py for a polished, printable HTML report with plain-language conclusions and minimum action packages.
  • Use Case: A team with 120 spare engineering hours must split capacity between a stable retention project and a risky AI experiment; the skill sizes both opportunities and returns a small test allocation with add, stop, and review conditions.

Quick Start

Ask the assistant to use the Kelly skill to decide how much of my budget or resource pool to allocate across my listed opportunities and produce a conservative allocation report.

Frequently Asked Questions about yao-kelly-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I calculate Kelly criterion position sizing for multiple opportunities?▼

Provide each opportunity's probabilities and scenario returns in a JSON brief, then run kelly_allocation_report.py. It computes standalone Kelly fractions first, then applies fractional Kelly, dependence haircuts, and a total exposure cap across the portfolio.

What is fractional Kelly and why use it instead of full Kelly?▼

Fractional Kelly multiplies the theoretical full Kelly fraction by a discount factor, such as 0.25 for medium confidence. It reduces drawdown risk when probability estimates are uncertain, which is why this skill defaults to fractional rather than full Kelly.

Can Kelly sizing be used for non-financial resources like engineering hours?▼

Yes, the skill supports any bounded resource pool through the resource_unit field, such as engineering-hours. The same sizing logic applies as long as downside is bounded and outcomes can be expressed as scenario returns.

When should I not use the Kelly criterion for a decision?▼

Avoid Kelly when the decision is irreversible, one-off, or has unbounded downside, and when probabilities cannot be approximated. The skill switches to a test-first or risk-review answer in those cases and never handles martingale escalation or guaranteed-return claims.

How do I generate an HTML report from Kelly sizing output?▼

Run generate_html_report.py with the JSON output from kelly_allocation_report.py as input. It renders a standalone printable HTML report with plain-language conclusions, action packages, scenario tables, and the round log.

Related Skills

Explore More Finance & Accounting Skills→
tjboudreauxtjboudreaux
941

thinking-bayesian

Apply Bayes' theorem to update beliefs with explicit priors and likelihoods.

Community
Advanced
systempromptiosystempromptio
2

marketing-hypothesis-ledger

Log marketing actions and hypotheses with H-### IDs in an append-only ledger.

Official
Advanced
thevibeworksthevibeworks
28

creating-financial-models

Create financial models with DCF, sensitivity analysis, and Monte Carlo simulation.

Official
Advanced
kerrybackkerryback

mean-variance-optimization

Compute tangency portfolio weights maximizing the Sharpe ratio from expected returns, risks, and correlations.

Community
Intermediate