Skip to main content
affpapa.org
RU EN
skill CX Codex

experiment-design — Claude Code skill: experiment planning

Turn a research idea into staged experiments with baselines, datasets, metrics, and ablations. Useful when shaping an experimental plan for a paper.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓ 
npx skills add lingzhi227/agent-research-skills --skill experiment-design -a claude-code
Open on GitHub

What experiment-design does

Give the skill a research idea, an existing plan, or a method description. It lays out four stages: get an implementation working, tune baselines, explore research improvements, then run ablation studies. Each stage builds on the strongest result from the one before it.

The design_experiments.py script prepares baseline comparisons, an ablation matrix, a hyperparameter grid, and metric choices. It can produce JSON or Markdown output. Python is required to run the script, which relies on the standard library only.

The plan includes datasets and completion criteria for each stage. During baseline tuning, the model architecture stays fixed and tests cover at least two datasets; the creative-research stage uses at least three. The skill also calls for multiple seeds and recording each run in notes.txt, so those steps belong in the experimental workflow.

What it helps with

  • break a research idea into staged experiments
  • list baseline methods for comparison
  • choose datasets and evaluation metrics
  • map out a hyperparameter grid
  • plan component-by-component ablations

Example prompts for the agent

  • Turn this research idea into a staged experiment plan.
  • Plan baseline tests and metrics for this classification method.
  • Draft ablation studies for the model described here.

How to install experiment-design

1Skills CLI (Vercel)

npx skills add lingzhi227/agent-research-skills --skill experiment-design -a claude-code

2Manually: put the files in place

~/.claude/skills/experiment-design/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/experiment-design/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about experiment-design

How do I install experiment-design in Claude Code?

Run npx skills add lingzhi227/agent-research-skills --skill experiment-design -a claude-code. The skill file is at ~/.claude/skills/experiment-design/SKILL.md.

What does experiment-design do?

It turns a research idea or method description into a staged experiment plan covering baselines, datasets, metrics, hyperparameters, and ablations.

Does experiment-design need Python?

Python is needed to run design_experiments.py. The script uses only the standard library.

Similar skills

SKILL.md skills catalog →
find-skills vercel-labs Discovers and installs agent skills from skills.sh directly inside your coding assistant. Handy when you need domain-specific workflows for a project. CX Codex +1 4,000,000installs frontend-design official anthropics This skill guides intentional visual design for new interfaces and redesigns. Use it when a generic generated UI is not enough. 1,025,790installs grill-me Matt Pocock grill-me interviews you about a plan or design to expose gaps and sharpen decisions. Use it before implementation starts. 1,424,476installs grill-with-docs Matt Pocock This skill interviews you about a plan or design and creates ADRs and a glossary along the way. Use it when decisions need sharper definition. 1,219,756installs improve-codebase-architecture Matt Pocock This skill scans a codebase for architectural friction, then builds a visual HTML report with refactoring candidates. Use it before a deeper architecture review. 1,163,035installs design-taste-frontend Leonxlnx This skill directs the design of public pages and work screens, then checks the result in a browser through screenshots. Use it when an interface feels generic or unfinished. CX Codex +1 678,700installs

From the same set30

data-analysis Writes statistical analysis code for a paper: picks a test that fits the data and reports p-values, effect sizes and confidence intervals. CX Codex +1 2,000installs deep-research Runs a systematic academic literature review on a topic: finds papers, reads them, builds a database and writes a final report. Useful when you need the state of a field, not a few search hits. CX Codex +1 1,900installs research-planning Turns a topic or idea into a structured research plan: methodology, paper sections, dependency-ordered tasks and diagrams. CX Codex +1 1,800installs github-research Finds open-source repositories matching research topics, inspects their codebases, and produces an integration blueprint. Useful when tracking down implementations of algorithms. CX Codex +1 1,800installs paper-revision Turns reviewer comments into concrete edits: each concern is tied to a section of the paper, then fixed. Handy once peer review comes back with revision requests. CX Codex +1 1,800installs paper-writing-section A skill that drafts or reworks a single section of an academic paper, from abstract to conclusion. Useful when the research is done but the prose is not. CX Codex +1 1,800installs novelty-assessment Runs a research idea through literature search and returns a verdict: novel or not. Useful before you commit to a direction. CX Codex +1 1,800installs idea-generation A skill that generates research ideas, refines them over several rounds and checks novelty against the literature. Useful before you commit to a paper topic. CX Codex +1 1,800installs

Project sponsors

Companies that keep this analytics open
Gold sponsor
Nutra.Club

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor