ab-test-analysis

ab-test-analysis is a command for Claude Code from lucasgaravelli/pm-skills-claude-code. It costs 0 tokens per session (2,660 once invoked), scanned A, a copy of ab-test-analysis, MIT.

An analysis workflow for A/B-test results, where two versions are compared to see whether one performs better. It checks statistical significance, meaning whether the observed difference is unlikely to be random, along with safety metrics.

In plain words
What is it for?
Use it to review sample sizes, conversions, test duration, guardrail metrics, and segments, then recommend shipping, investigating, extending, or stopping the test.
Why use it?
It helps distinguish a real improvement from noise and makes the result easier to explain to people who do not work with statistics.

Command for Claude Code

Written for Claude Code: $ARGUMENTS substitution.

Good fit Use it to review sample sizes, conversions, test duration, guardrail metrics, and segments, then recommend shipping, investigating, extending, or stopping the test.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/lucasgaravelli/pm-skills-claude-code/ab-test-analysis
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/lucasgaravelli/pm-skills-claude-code

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ab-test-analysis

README.md
[![agentmods](https://agentmods.dev/badge/commands/lucasgaravelli/pm-skills-claude-code/ab-test-analysis/github.svg)](https://agentmods.dev/commands/lucasgaravelli/pm-skills-claude-code/ab-test-analysis)
Your own site
<a href="https://agentmods.dev/commands/lucasgaravelli/pm-skills-claude-code/ab-test-analysis"><img src="https://agentmods.dev/badge/commands/lucasgaravelli/pm-skills-claude-code/ab-test-analysis/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ab-test-analysis

Your own site · 80×15
<a href="https://agentmods.dev/commands/lucasgaravelli/pm-skills-claude-code/ab-test-analysis"><img src="https://agentmods.dev/badge/commands/lucasgaravelli/pm-skills-claude-code/ab-test-analysis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,660 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.02660
Opus 5 $0.00000 $0.01330
Sonnet 5 $0.00000 $0.00532
Haiku 4.5 $0.00000 $0.00266

Measured 9d ago against content hash 86bb8bb43ab7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

ab-test-analysis scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

100% identical to ab-test-analysis — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.claude/commands/ab-test-analysis.md · 302 lines

How it starts

The opening of the file, as written. The whole thing — 302 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/ab-test-analysis

O que essa skill faz

Analisa resultados de A/B tests com rigor estatístico: calcula significância, interpreta resultados, checa guardrail metrics e fornece recomendação clara de ação (ship / investigar / estender / parar).

Saída: Analysis Summary com tabela de resultados, interpretação estatística e recomendação.


Quando usar

  • Experimento A/B terminou e precisa analisar resultados
  • Quer validar se diferença observada é real ou ruído
  • Precisa comunicar resultados para stakeholders não-técnicos
  • Quer verificar se guardrail metrics foram impactadas
  • Teste está rodando e quer fazer interim check (com cuidado)

Input esperado

Mínimo:

  • Hipótese: O que esperava acontecer e por quê
  • Variantes: Descrição do control e treatment
  • Primary metric: Qual métrica principal de sucesso
  • Dados: Tamanho da amostra e conversão por variante
  • Duração: Quanto tempo o teste rodou

Opcional:

  • Guardrail metrics e seus valores
  • Segmentação (device, geo, user type)
  • Baseline histórico da métrica
  • MDE (Minimum Detectable Effect) planejado
  • Split ratio (se diferente de 50/50)

Processo de análise

Passo 1 — Entender o experimento

Antes de olhar números, esclarecer:

EXPERIMENT OVERVIEW
═══════════════════════════════════════════════════

Hipótese: [O que esperava e por quê]
Control (A): [descrição]
Treatment (B): [descrição]
Primary metric: [nome e definição]
Guardrail metrics: [lista]
Duração: [X dias]
Split: [% control / % treatment]
Trigger: [quem foi incluído no teste]

Passo 2 — Validar setup

Checklist de sanidade antes de confiar nos resultados:

  • Sample size adequado? N suficiente para detectar o MDE desejado
  • Duração cobre ciclos de negócio? Mínimo 1-2 semanas, incluindo dia útil + fim de semana
  • Randomização correta? Distribuição uniforme entre variantes
  • Sample Ratio Mismatch (SRM)? Verificar se split real ≈ split planejado
  • Novelty effect descartado? Usuários existentes podem reagir à mudança, não à melhoria
  • Peeking controlado? Resultados não foram analisados múltiplas vezes sem correção

Read the full file on GitHub · 302 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 302 lines · 0 tokens per session scan A 86bb8bb43ab7

Subscribe to this mod's changes

ab-test-analysis is a command published in the GitHub repository lucasgaravelli/pm-skills-claude-code (20 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,660 tokens. A static security scan graded it A with 0 findings. It is 100% identical to ab-test-analysis, differing in 0 lines, and is treated as a copy.