skills.rest
AndurilCode
AndurilCodeOFFICIAL
Explore all skills by this author →
GitHub

probabilistic-thinking

Convert vague claims into Bayesian probability estimates with base rates and expected values.

7|2|Updated Mar 5, 2026
Search Tags:#decision making#uncertainty#risk assessment#base rates#expected value#bayesian updating
One-click install
npx skills add https://github.com/AndurilCode/craftwork --skill probabilistic-thinking
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: probabilistic-thinking
Source: https://github.com/AndurilCode/craftwork/tree/main/skills/probabilistic-thinking
Command: npx skills add https://github.com/AndurilCode/craftwork --skill probabilistic-thinking

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you replace vague confidence and “probably” judgments with explicit probability estimates, so you can reason clearly under uncertainty, compare risks, and update beliefs when new evidence arrives.

Core Features & Use Cases

  • Base-rate reasoning: checks reference-class likelihoods to avoid common base-rate errors when estimating how often an event type occurs.
  • Bayesian updating: updates beliefs using priors and evidence-weighting so estimates move appropriately with stronger vs. weaker information.
  • Decision support via expected value: compares options using EV (probability × payoff) while accounting for both upside and downside, not just the best-case outcome.
  • Calibration & uncertainty hygiene: prompts users to use confidence ranges and flags issues like treating uncertain outcomes as certainties, anchoring, and overreacting to recent data.

Quick Start

Ask: "Given these facts, estimate the probability of X, state the base rate you used, perform a Bayesian update with the new evidence, and compare options A vs B by expected value."

Frequently Asked Questions about probabilistic-thinking

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

FAQPage Schema
How do I estimate probability using base rates and Bayesian updating?▼

To estimate probability using base rates, you first establish a reference-class base rate as your prior, then perform a Bayesian update by weighting new evidence to shift the probability accordingly. This prevents overreacting to weak signals.

How does expected value calculation improve risk assessment under uncertainty?▼

Expected value calculation improves risk assessment by multiplying probability by payoff to compare options quantitatively. This accounts for both upside and downside outcomes rather than focusing solely on best-case scenarios.

What is the best way to calibrate uncertainty and avoid reasoning errors in decision making?▼

The best way to calibrate uncertainty is to replace vague confidence with explicit probability ranges and check reference-class likelihoods. This flags issues like anchoring, treating uncertain outcomes as certainties, and ignoring base rates.

Can I use Bayesian updating for decision making with incomplete information?▼

Yes, Bayesian updating is specifically designed for decision making with incomplete information. It turns vague claims into probabilistic beliefs by applying likelihood and evidence rationale to update your prior probabilities.

What are common base rate errors in probability estimation and how do I avoid them?▼

Common base rate errors occur when you ignore reference-class likelihoods and overreact to specific signals. You avoid them by explicitly stating base rates and treating uncertain outcomes with confidence ranges rather than certainties.

Related Skills

Explore More Product & Management Skills→
jacob-balslevjacob-balslev
1

bayesian-reasoning

Update posterior confidence from priors and evidence strength under uncertainty.

Community
Intermediate
lyndonkllyndonkl
142

expected-value

Calculate expected value from probabilities and payoffs for uncertain decisions.

Community
Advanced
PatrickSUDOPatrickSUDO
136

ev-check

Calculate expected value and probability distribution for investment portfolios.

Community
Advanced
danielsimonjrdanielsimonjr
1

think-probabilistic

Compute posterior beliefs using Bayesian updating and Dempster-Shafer reasoning.

Community
Advanced