About this skill

Problem

In hiring, offer selection, apartment hunting, car buying, and fund screening, the difficulty is often not a lack of options but a lack of a stopping rule: keep searching and risk missing a good option, or decide too early and regret it. smart-chooser turns these sequential-choice problems into a computable process using the 37% optimal stopping rule to separate observation from decision-making.

How It Works

The skill collects candidate options and decision dimensions such as price, commute, team, salary, quality score, or budget, and supports positive and inverse metrics. It normalizes heterogeneous data to 0-1, applies user-defined weights, and produces a 0-100 composite score. Key steps include:
- Information collection: identify options, dimensions, weights, and constraints.
- Scoring: use scripts such as score.js and pipeline.js to rank candidates.
- 37% filtering: compute the observation cutoff with ceil(N * 0.3679), record the best standard during the early segment, and select once a later option exceeds that standard.
- Strategy comparison: support 37, greedy, satisficing, and explore strategies.
- Reporting: output rankings, observation analysis, a recommended choice, execution checklist, and optionally an HTML report.

Boundaries

This skill is best for one-time, sequential, hard-to-reverse choices. It is less suitable for fully reversible, fully informed, or highly stochastic situations. The observation phase must be strictly observe only, and the decision phase should stop searching once the threshold is met. It provides a decision framework and calculation support, not business judgment or investment advice.

Screenshots

Use Cases

Rank 15 product-manager interview candidates by weighted criteria and identify when to start hiring after the 37% observation phase.
Score 10 apartments on rent, commute, and size, compute the first 4 as the observation set, and recommend when to sign.
Compare 20 funds by risk, return, and fees, then contrast recommendations from 37%, greedy, and satisficing strategies.
Evaluate 30 article topics by appeal, cost, and fit, and use the 37% rule to decide when to start writing.

Best For

Hiring managers who need to score candidates on weighted interview criteria and determine when to move from observation to hiring.
Rent or car shoppers who want to quantify multi-dimensional options and get a defensible stopping point.
Content editors comparing topic appeal, cost, and fit to decide when to start drafting.
Fund screeners who need weighted ranking by risk, return, and fees and want to compare decision strategies.