experiment-workspace

Create, update, and archive database-backed experiments with Bayesian A/B testing.

1|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/featbit/featbit-release-decision-agent --skill experiment-workspace
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: experiment-workspace
Source: https://github.com/featbit/featbit-release-decision-agent/tree/main/skills/experiment-workspace
Command: npx skills add https://github.com/featbit/featbit-release-decision-agent --skill experiment-workspace

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires project-sync, scripts/analyze.ts, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill centralizes the creation, analysis, and management of experiments by replacing typical dashboard workflows with database-backed records, enabling seamless tracking and decision-making.

Core Features & Use Cases

  • Experiment Management: Automates creation, updating, and archiving of experiments directly in the database.
  • Analysis Orchestration: Triggers server-side Bayesian or bandit analyses, integrating results into the workflow.
  • Use Case: A product team wants to run an A/B test, analyze data with Bayesian methods, and automatically plan the next steps based on the outcome, all within their codebase.

Quick Start

Use this Skill to create and analyze a new experiment by providing the flag details and desired metrics.

Frequently Asked Questions about experiment-workspace

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

FAQPage Schema
How do I track A/B testing experiments directly within a database environment?

You can track A/B testing experiments in a relational database by using an integrated system that replaces dashboard workflows with database-backed records. This enables seamless experiment lifecycle management from setup through analysis and closure.

What is the best way to run Bayesian analysis for product experiments without leaving my codebase?

The best way to run Bayesian analysis within your codebase is to use an integrated experiment management system that triggers server-side analysis endpoints. This orchestrates Bayesian A/B testing and automatically plans next steps based on the outcome.

How do I manage the full experiment lifecycle from setup to closure automatically?

You manage the full experiment lifecycle by defining, tracking, and analyzing experiments through a database-backed system. This ensures data integrity and decision traceability from initial setup through analysis to final closure.

Do I need a relational database to use bandit algorithms for experiment tracking?

Yes, you need synchronization with a relational database and server-side analysis endpoints to effectively use bandit algorithms for experiment tracking. This setup ensures data integrity and decision traceability for your experiments.

When should I not use a database-backed workflow for A/B testing?

You should avoid a database-backed workflow for A/B testing if you lack a relational database or server-side analysis endpoints. This system requires synchronization with these components to function effectively and ensure data integrity.