experiment-plan

Plan claim-driven experiments with phases, run orders, and evaluation protocols.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/KwongFuk/codex-skills --skill experiment-plan-kwongfuk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: experiment-plan
Source: https://github.com/KwongFuk/codex-skills/tree/main/global/experiment-plan
Command: npx skills add https://github.com/KwongFuk/codex-skills --skill experiment-plan-kwongfuk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Turn refined research proposals into a concrete, claim-driven experiment roadmap that supports paper-ready validation.

Core Features & Use Cases

  • Phase-based planning blocks including main anchor, novelty isolation, simplicity check, frontier necessity check, and failure analysis.
  • Clear run order and computation budget planning, plus baseline selection.
  • Workflow-driven outputs: an experiment plan doc and an experiment tracker.

Quick Start

Provide your research proposal and constraints, then ask for a detailed experiment plan.

Frequently Asked Questions about experiment-plan

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

FAQPage Schema
How do I structure an experiment plan for paper-ready validation?

Create an experiment plan by defining phase-based blocks, run orders, compute budgets, and evaluation protocols from your research proposal. This produces a structured experiment plan doc and tracker for publication-ready validation.

What is claim-driven experiment design in machine learning research?

Claim-driven experiment design isolates novelty and tests specific hypotheses through structured run orders and baselines. It transforms research proposals into targeted phase blocks like anchor tests and failure analysis for rigorous validation.

How do I plan compute budgets and run orders for ML experiments?

Plan compute budgets by mapping phase-based experiment blocks to resource constraints and defining a sequential run order. This ensures systematic baseline selection and execution tracking across your research validation workflow.

Can I use this experiment planning workflow for non-ML research domains?

Yes, this experiment planning workflow applies across research domains, defining success criteria, baselines, and documented execution steps for publication-ready validation. It structures any proposal into claim-driven evaluation protocols.

What's the best way to define baselines and success criteria for research validation?

Define baselines and success criteria by breaking proposals into main anchor, novelty isolation, and simplicity check blocks. This phase-based approach documents clear evaluation protocols and execution steps for rigorous paper validation.

Why do I need a failure analysis phase in my experiment plan?

A failure analysis phase identifies edge cases and validates frontier necessity within your experiment plan. It ensures your claim-driven validation addresses potential weaknesses before publication, strengthening overall research design rigor.