MarketingPROPRO REQUIREDFC-MKT-012
A/B Test Claw
ab-test-claw
v0.2.0May 22, 2026A/B tests with statistical rigor — power analysis before you ship, significance checks, and effect size estimates
Most A/B tests ship noise as signal. I care about statistical power before you launch (minimum detectable effect, sample size calculations with base-rate awareness), the peeking problem (if you check your dashboard daily, your p-value is a lie), segment-level Simpson's paradox where the aggregate says "variant B wins" but every segment says "variant A wins," and the revenue-vs-significance trade-off when p=0.06 but the ARPU lift is $4. I generate test designs with power analysis, preregistration templates, and sequential testing protocols with alpha-spending functions.
OpenClaw
Hermes
Claude Code
Codex
Cursor
Windsurf
Aider
OllamaPRIMARY ACTION
Unlock with ProCOMPATIBLE WITH
OpenClawHermesClaude CodeCodex+4
8 compatible tools4 output formats
When to Use
- Improve technical SEO and indexability
- Create launch messaging for technical users
- Strengthen conversion pages with better structure
- Translate features into clear value
Compatible Frameworks
OpenClawHermesClaude CodeCodexCursorWindsurfAiderOllama8 TOOLS
Quality Gates
- Testable and specific hypothesis
- Primary and guardrail metrics defined
- Sample size calculated and justified
- Pre-registered analysis plan
- No statistical pitfalls present
4 GATES DEFINED
Expected Outputs
A/B Test Design Document with Power Analysis and Sample Size CalculationPreregistration Template with Primary Metric, Guardrail Metrics, and Stopping RuleSequential Testing Protocol with Alpha-Spending Function and Interim Analysis SchedulePost-Test Analysis with Segment-Level Simpson's Paradox Check and Revenue Impact
Native exports per tool
OpenClaw10 files
openclaw/AGENTS.mdopenclaw/SOUL.mdopenclaw/TOOLS.md+7 moreHermes5 files
hermes/skills/flickclaw/ab-test-claw/SKILL.mdhermes/skills/flickclaw/ab-test-claw/references/workflow.mdhermes/skills/flickclaw/ab-test-claw/references/quality-gates.md+2 moreClaude Code6 files
claude-code/CLAUDE.mdclaude-code/.claude/skills/ab-test-claw/SKILL.mdclaude-code/.claude/skills/ab-test-claw/references/workflow.md+3 moreCodex5 files
codex/AGENTS.mdcodex/.flickclaw/agents/ab-test-claw/codex.mdcodex/.flickclaw/agents/ab-test-claw/workflow.md+2 moreCursor3 files
cursor/.cursor/rules/flickclaw-ab-test-claw.mdccursor/.cursor/rules/flickclaw-ab-test-claw-workflow.mdccursor/.cursor/rules/flickclaw-ab-test-claw-quality-gates.mdcWindsurf3 files
windsurf/.windsurf/rules/flickclaw-ab-test-claw.mdwindsurf/.windsurf/rules/flickclaw-ab-test-claw-workflow.mdwindsurf/.windsurf/rules/flickclaw-ab-test-claw-quality-gates.mdAider3 files
aider/CONVENTIONS.mdaider/aider.mdaider/.aider.conf.ymlOllama4 files
ollama/Modelfileollama/system-prompt.mdollama/template.md+1 moreInstall Commands
Install the FlickClaw CLI, then select your AI agent framework below to get the correct install command.
Step 1: Install CLI (one-time)
npm install -g @flickclaw/cli@latestStep 2: Select Framework
OpenClaw
npm exec --yes @flickclaw/cli@latest -- install ab-test-claw --target openclawDownload as ZIP
Example Prompt
Try this prompt with A/B Test Claw to see what it can do:
Create a launch campaign for the new feature. Include audience analysis, messaging strategy, and A/B Test Design Document with Power Analysis and Sample Size Calculation, Preregistration Template with Primary Metric, Guardrail Metrics, and Stopping Rule, Sequential Testing Protocol with Alpha-Spending Function and Interim Analysis Schedule.Example Output
IllustrativeWhat a typical A/B Test Claw report looks like:
# A/B Test Claw — Assessment Report **Project**: saas-landing **Context**: a SaaS product landing page targeting engineering leads **Generated**: 2026-07-10 ## Executive Summary The A/B Test Claw completed its review of saas-landing (a SaaS product landing page targeting engineering leads). 3 findings were identified with concrete remediation steps. All quality gates were verified before delivery. ## Findings | # | Severity | Area | Finding | Recommended Action | |---|----------|------|---------|-------------------| | 1 | **P1** | SEO | Missing meta description on feature pages | Add unique 150-char descriptions with primary keyword | | 2 | **P1** | Copy | Hero copy focuses on features, not outcomes | Rewrite hero around time savings and error reduction | | 3 | **P2** | Conversion | CTA below fold on mobile | Add sticky CTA or move above fold | ## Quality Gates - [✓] statistical_power_gte_80_percent - [✓] simpsons_paradox_segment_check - [✓] peeking_correction_or_sequential_protocol ## Outputs Generated - **A/B Test Design Document with Power Analysis and Sample Size Calculation**: Included in the report above. - **Preregistration Template with Primary Metric, Guardrail Metrics, and Stopping Rule**: Included in the report above. - **Sequential Testing Protocol with Alpha-Spending Function and Interim Analysis Schedule**: Included in the report above. - **Post-Test Analysis with Segment-Level Simpson's Paradox Check and Revenue Impact**: Included in the report above. ## Validation - [x] All quality gates passed (3/3) - [x] 3 findings documented with severity and remediation - [x] 4 output sections generated - [x] Evidence collected and referenced --- *This is an illustrative example output from FlickClaw. Results vary based on project context.*
RELATED AGENTS