What problem does it solve?
It turns vague, high-uncertainty choices into a structured, auditable decision report with explicit priors, graded evidence, posterior updates, action thresholds, and sensitivity checks, instead of relying on gut feel or unstructured brainstorming.
Core Features & Use Cases
- Bayesian Decision Pipeline: Structure a decision into hypothesis, time horizon, success metric, and actions; set a prior, grade evidence, update the posterior via odds or Beta-binomial methods, and compare action thresholds by expected value.
- Multi-Turn Decision Loop: Start from incomplete input with a weak prior, ask the minimum next questions, and log each round's probability change, readiness, and remaining gaps.
- Automated Report Export: Generate a Chinese-first Markdown report plus a bilingual (Chinese/English) HTML report with sticky navigation, collapsible sections, and print/save-as-PDF support.
- Use Case: A product team unsure whether to launch a new feature this quarter feeds in interview signals, pilot data, and cost parameters, then receives a report recommending a paid pilot first, with sensitivity analysis showing when the recommendation would flip.
Quick Start
Ask the assistant to use the yao-bayesian-skill to evaluate whether you should proceed with a specific decision, providing your hypothesis, available evidence, and possible actions.
