experiment-analyzer

experiment-analyzer is a skill for Claude Code from reatlat/fullstory-claude-plugin. It costs 52 tokens per session (1,218 once invoked), scanned A, original, MIT.

An experiment-analysis workflow for comparing two or more versions of something, such as a checkout page. It checks which version performed better, how large the difference was, and whether it may be random variation.

In plain words
What is it for?
Use it to analyze A/B tests, compare results before and after a change, examine user groups such as device types, and check side effects such as more errors.
Why use it?
It helps turn experiment results into a decision instead of relying only on raw totals or impressions.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the fullstory-claude-plugin plugin — 46 skills, 3 agents, 1 MCP server shipped together

Good fit Use it to analyze A/B tests, compare results before and after a change, examine user groups such as device types, and check side effects such as more errors.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/reatlat/fullstory-claude-plugin/experiment-analyzer
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.

Any agent
npx skills add reatlat/fullstory-claude-plugin --skill experiment-analyzer
Clone the repo
git clone --depth 1 https://github.com/reatlat/fullstory-claude-plugin

Made for: Claude Code.

Or install fullstory-claude-plugin, the plugin that ships this one along with the rest of its 46 skills, 3 agents, 1 MCP server.

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 experiment-analyzer

README.md
[![agentmods](https://agentmods.dev/badge/skills/reatlat/fullstory-claude-plugin/experiment-analyzer/github.svg)](https://agentmods.dev/skills/reatlat/fullstory-claude-plugin/experiment-analyzer)
Your own site
<a href="https://agentmods.dev/skills/reatlat/fullstory-claude-plugin/experiment-analyzer"><img src="https://agentmods.dev/badge/skills/reatlat/fullstory-claude-plugin/experiment-analyzer/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 experiment-analyzer

Your own site · 80×15
<a href="https://agentmods.dev/skills/reatlat/fullstory-claude-plugin/experiment-analyzer"><img src="https://agentmods.dev/badge/skills/reatlat/fullstory-claude-plugin/experiment-analyzer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,218 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found 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 $0.00052 $0.01218
Opus 5 $0.00026 $0.00609
Sonnet 5 $0.00010 $0.00244
Haiku 4.5 $0.00005 $0.00122

Measured 10d ago against content hash 546ccd8f0fe3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

experiment-analyzer 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 10d 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.

skills/experiment-analyzer/SKILL.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Experiment Analyzer

Analyze A/B tests and experiments — measure impact, check statistical validity, break down results by segment, and tell you which variant won (and by how much).

When to Use

  • "Did the new checkout design improve conversion?"
  • "Analyze the 'blue button vs green button' experiment"
  • "Is the pricing page experiment statistically significant?"
  • "Break down the experiment results by device type"
  • "Which variant performed better for enterprise users?"
  • "Should we ship the new onboarding flow?"

Mental Model

An experiment compares two or more variants against a control. The goal is to determine:

  1. Direction: Which variant performed better?
  2. Magnitude: By how much? (+5%? +20%?)
  3. Confidence: Is the difference real or noise? (statistical significance)
  4. Segments: Does it work for everyone, or only certain users?
  5. Side effects: Did the variant improve conversion but increase errors?

Workflow

Step 1: Define the experiment

Ask the user:

  • What's being tested? (e.g., checkout flow redesign)
  • What's the success metric? (e.g., checkout completion rate)
  • When did the experiment start? (so you can scope the time range)
  • How are users split? (50/50 random? By user property? By feature flag?)

If users are split by a user property (e.g., experiment_group = "control" or "variant"), use segments. If split by a feature flag that maps to a custom event, filter by that event.

Step 2: Build metrics per variant

For user-property splits (experiment_group property):

fullstory:build_segment("users in experiment group 'control'") → seg_control
fullstory:build_segment("users in experiment group 'variant'") → seg_variant
fullstory:build_metric(query="checkout completion rate", output_type="single_number")
fullstory:update_metric(metric_id, segment_id=seg_control)
fullstory:compute_metric(metric_id) → control_result
fullstory:update_metric(metric_id, segment_id=seg_variant)
fullstory:compute_metric(metric_id) → variant_result

Read the full file on GitHub · 118 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. 10d ago First seen · 118 lines · 52 tokens per session scan A 546ccd8f0fe3

Subscribe to this mod's changes

experiment-analyzer is a skill published in the GitHub repository reatlat/fullstory-claude-plugin (62 stars, last pushed 28d ago), licensed MIT. It adds 52 tokens to every session and 1,218 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.