Can this skill help with software product development?

Yes, it helps teams avoid copying features from successful competitors without understanding if those same features also existed in products that failed for other reasons.

In what domains is this skill most useful?

It is highly effective in business strategy, financial analysis, startup consulting, historical research, and any data science project involving non-random sampling.

How does this skill change Claude's behavior?

It prompts Claude to ask 'trigger questions' about what's missing from a dataset and to actively seek out negative evidence before validating a 'success pattern' or strategic advice.

What is the primary goal of the Survivorship Bias skill?

The goal is to prevent the model from drawing overly optimistic or false conclusions by forcing it to account for 'missing' data—the entities that failed to pass a selection process and are no longer visible.

Survivorship Bias Thinking Pattern

Name: Survivorship Bias Thinking Pattern
Author: lev-os

bylev-os

0•

Data Science & ML

Enables Claude to identify and mitigate logical errors caused by focusing on visible successes while ignoring hidden failures.

The Survivorship Bias skill provides a rigorous framework for Claude to evaluate datasets and narratives where failures are systematically excluded. By implementing the seven-step process derived from Abraham Wald's WWII insights, this skill helps users avoid false patterns in business strategy, investment analysis, and product development. It forces the consideration of the 'invisible' data—the startups that failed, the products that were discontinued, and the funds that closed—ensuring that conclusions are based on a representative population rather than just the winners.

Key Features

01Detects anti-patterns in success-based advice and 'best practice' lists

02Identifies selection mechanisms that make failures invisible in datasets

03Calculates conditional probabilities to distinguish between correlation and causation

04Guides prospective analysis to track cohorts forward rather than retrospectively

05Provides a structured process for seeking out failure-case data and postmortems

060 GitHub stars

Use Cases

01Critiquing business case studies to separate true success drivers from survivor luck

02Reviewing historical data or investment performance for selection bias

03Designing research methodologies that capture negative results and failed experiments

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add lev-os/agents survivorship-bias

For use in Claude.ai and ChatGPT

Download Skill