feature-metrics

Define feature success metrics using the STEDII framework with baselines, guardrails, and kill criteria.

1|Updated Jul 25, 2026
One-click install
npx skills add https://github.com/zacgoodwin/AIBootstrap --skill feature-metrics-zacgoodwin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: feature-metrics
Source: https://github.com/zacgoodwin/AIBootstrap/tree/main/.claude/skills/feature-metrics
Command: npx skills add https://github.com/zacgoodwin/AIBootstrap --skill feature-metrics-zacgoodwin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Product managers often pick vague or vanity metrics that cannot detect whether a feature actually worked. This Skill guides you through selecting a primary metric, guardrail metrics, and kill criteria that pass the six STEDII criteria (Sensitive, Timely, Easy to understand, Directional, Implementable, Independent), grounded in your existing PRDs, business context, and historical baselines. ## Core Features & Use Cases - STEDII Metric Validation: Evaluates every proposed metric against six criteria with explicit pass/fail reasoning. - Context-Aware Baselines: Automatically pulls feature context from PRDs, business info, historical metrics, strategy docs, and meeting notes before proposing targets. - Structured Output Template: Produces a metrics definition with primary metric, baseline, target, guardrails with acceptable ranges, kill criteria, and a measurement plan. - Use Case: Before launching an A/B test for a new onboarding flow, run this Skill to define activation rate as the primary metric, set page load time and error rate as guardrails, and establish rollback thresholds. ## Quick Start Ask the assistant to define success metrics for your feature using the STEDII framework, providing the feature name, the user behavior it changes, and the expected business outcome.

Frequently Asked Questions about feature-metrics

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

FAQPage Schema
How do I define success metrics for a new feature?

Identify the user behavior the feature changes and the business outcome expected, then select one primary metric that passes the STEDII criteria: Sensitive, Timely, Easy to understand, Directional, Implementable, and Independent. Add guardrail metrics with acceptable ranges and kill criteria for early rollback.

What is the STEDII framework for experiment metrics?

STEDII is a six-criteria checklist for trustworthy metrics: Sensitive (detects change), Timely (responds within the experiment window), Easy to understand, Directional (clear good/bad direction), Implementable (data is collectible), and Independent (isolated from external factors like seasonality).

What is the difference between primary metrics and guardrail metrics?

A primary metric is the single measure that defines success and drives the go/no-go decision. Guardrail metrics are not goals; they protect against unintended harm, such as page load time, error rate, or support tickets, and are monitored against acceptable ranges rather than targets.

How do I choose a primary metric for an A/B test?

Pick one metric directly tied to the feature goal with a known baseline, a specific numeric target, and a timeline. Verify it is sensitive enough to detect the expected impact given your traffic and variance, and confirm it ladders up to your company's North Star metric.

When should I use kill criteria in an experiment?

Define kill criteria before launching any experiment. They specify thresholds, such as a metric dropping below a floor or a qualitative safety signal, that trigger immediate rollback to prevent serious negative impact on users or the business.