Experiment Craft

When your experiment isn't working and you're stuck running random variations hoping something sticks, this skill forces you into a structured 5-step diagnostic: collect failure cases, find a working baseline (even if you have to radically simplify), incrementally bridge the gap to isolate what breaks it, verify hypotheses with targeted experiments, then fix the actual cause. It's built around the idea that understanding why experiments fail matters more than running more of them. Comes with experiment logging templates that track purpose, setting, results, analysis, and critically, your own next steps. Works standalone or gets called from experiment-pipeline when a stage fails. The counterintuitive rules section is worth reading even if you don't use the whole flow.

skills.sh installs: 380
GitHub stars: 488
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add evoscientist/evoskills --skill experiment-craft --agent claude-code

Installs into .claude/skills of the current project.

Files

Experiment Craft

A systematic approach to running, debugging, and iterating on research experiments. The critical skill is not running more experiments — it's understanding WHY experiments fail.

When to Use This Skill

User's experiment is not working or producing unexpected results
User needs help diagnosing why a method fails on certain data
User wants to organize their experiment process with structured logging
User asks about debugging research code or iterating on approaches
User mentions "experiment debugging", "why doesn't this work", "experiment log", "results are wrong"

This skill is typically loaded from within experiment-pipeline when a stage attempt fails. After debugging, return to the pipeline's stage-gate structure to continue. Can also be used standalone for any experiment debugging.

The Debugging Mindset

Finding WHY experiments fail is the most critical research skill. Not analyzing results leads to two failure modes:

Slow progress: Running random experiments without understanding failure causes
Wasted time: Abandoning good approaches because activation tricks were missed

SKILL.md on GitHub