accelerate-learn

Classify completed A/B test experiments and update a local learning journal.

8|Updated Apr 14, 2026
One-click install
npx skills add https://github.com/humanmade/accelerate-ai-toolkit --skill accelerate-learn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: accelerate-learn
Source: https://github.com/humanmade/accelerate-ai-toolkit/tree/main/skills/accelerate-learn
Command: npx skills add https://github.com/humanmade/accelerate-ai-toolkit --skill accelerate-learn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill reads the site's completed A/B test results, classifies each experiment into a canonical suggestion pattern, and updates the site's learning journal so every other skill can tailor future recommendations to what has worked here.

This skill is read-only on WordPress. You never create, modify, or stop experiments. You write two local files only.

Core Features & Use Cases

  • Fetch site context and list completed experiments to build a complete picture of site experiments.
  • Classify experiments by pattern and compute outcomes to populate a centralized learning journal that informs future recommendations.
  • Update the local journal files to enable other skills to tailor suggestions based on historical results.

Quick Start

Fetch recent A/B results, classify experiments into patterns, and update the local learning journal.

Frequently Asked Questions about accelerate-learn

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

FAQPage Schema
How do I use completed A/B test results to improve future recommendations?

You can use completed A/B test results to improve recommendations by classifying experiments into canonical suggestion patterns and updating a local learning journal to reflect what works best on your site.

Can I analyze A/B test experiments on WordPress without modifying live tests?

Yes, you can analyze A/B test experiments on WordPress in a read-only mode. The process fetches site context and lists completed experiments without creating, modifying, or stopping any active tests.

How does a learning journal apply pattern classification to A/B testing outcomes?

A learning journal applies pattern classification by sorting completed A/B test experiments into canonical suggestion patterns, computing outcomes, and merging these results into a centralized state to inform future recommendations.

What is the best way to centralize historical A/B test data for other tools to access?

The best way to centralize historical A/B test data is to write results into local journal files. This private state allows other skills to access the historical outcomes and tailor future suggestions accordingly.

Does analyzing A/B test results write data directly to my WordPress site?

No, analyzing A/B test results does not write data to WordPress. The operation is strictly read-only on the platform and only writes two local journal files privately to record the classified outcomes.