Ablation Planner

wanshuiyin/auto-claude-code-research-in-sleep

You've got main results that support your claims, and now you need ablations before submission. This skill uses Codex to design the ablation studies from a reviewer's perspective, asking what components actually matter and what alternatives you should have tried. It generates a prioritized list with clear hypotheses for each ablation, then helps you implement them in a sensible order. The division of labor is clean: Codex thinks like a skeptical reviewer, you handle feasibility and compute budget. It's explicitly designed to prevent busywork ablations that won't add insight. Best used when you're past the "does it work" phase and need to show which parts actually contribute to your results.

skills.sh installs: 537
GitHub stars: 18.9K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add wanshuiyin/auto-claude-code-research-in-sleep --skill ablation-planner --agent claude-code

Installs into .claude/skills of the current project.

Files

Ablation Planner

Systematically design ablation studies that answer the questions reviewers will ask. Codex leads the design (reviewer perspective), CC reviews feasibility and implements.

Context: $ARGUMENTS

When to Use

Main results pass /result-to-claim with claim_supported = yes or partial
User explicitly requests ablation planning
/auto-review-loop reviewer identifies missing ablations

Workflow

Step 1: Prepare Context

CC reads available project files to build the full picture:

Method description and components (from idea-stage/docs/research_contract.md, legacy docs/research_contract.md, or project CLAUDE.md)
Current experiment results (from EXPERIMENT_LOG.md, EXPERIMENT_TRACKER.md, or W&B)
Confirmed and intended claims (from result-to-claim output or project notes)
Available compute resources (from CLAUDE.md server config, if present)

Step 2: Codex Designs Ablations

mcp__codex__codex:
  model: gpt-6-astra
  config: {"model_reasoning_effort": "xhigh"}
  prompt: |
    You are a rigorous ML reviewer planning ablation studies.
    Given this method and results, design ablations that:

SKILL.md on GitHub