Auto Experiment

Auto Experiment is an Automation skill for Claude Code, published by Xiangyue-Zhang in auto-deep-researcher-24x7.

1.3K stars111 forkson Xiangyue-Zhang/auto-deep-researcher-24x7Added 2026/08/20Repository updated 2026/06/03

ai-agentautonomous-agentclaude-codedeep-learningexperiment-automationgpuhyperparameter-tuningllm-agentmachine-learningmlopspytorchresearch-automation

Install in seconds

Install Auto Experiment

Copy Auto Experiment into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/Xiangyue-Zhang/auto-deep-researcher-24x7/tree/main/skills/auto-experiment ~/.claude/skills/auto-experiment

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/Xiangyue-Zhang/auto-deep-researcher-24x7.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/auto-experiment/SKILL.md in Xiangyue-Zhang/auto-deep-researcher-24x7
Installs to: ~/.claude/skills/auto-experiment
Collection: One of 8 skills cataloged from this repository
Category: Automation — 1523 skills

What Auto Experiment does

Auto Experiment launches an autonomous THINK→EXECUTE→REFLECT loop that runs deep learning experiments 24/7. Use it to iterate on GPU training, monitor runs at zero LLM cost, and automatically decide next steps from results.

Auto Experiment is cataloged under Automation on DirSkills. Auto Experiment comes from a repository tagged ai-agent, autonomous-agent, claude-code, deep-learning and experiment-automation.

Documentation

README

Excerpt

auto-experiment

Launch an autonomous experiment agent that runs your deep learning experiments 24/7.

What This Does

This skill starts a THINK → EXECUTE → REFLECT loop that:

Reads your PROJECT_BRIEF.md to understand the research goal
Analyzes previous results in MEMORY_LOG.md
Plans the next experiment (hypothesis + success criteria)
Implements code changes and runs a mandatory dry-run
Launches GPU training via nohup (tracks PID)
Monitors at zero LLM cost (only kill -0 PID + tail log + nvidia-smi)
Wakes up when training finishes to analyze results
Updates memory and decides: iterate, pivot, or report
Repeats

Usage

This is the opening of the README. Read the full README on GitHub.

Commands Auto Experiment provides

Slash commands named in this skill’s SKILL.md, listed in the order they first appear.

/auto-experiment

Frequently asked about Auto Experiment

What else does Xiangyue-Zhang publish alongside Auto Experiment?
Auto Experiment is one of 8 skills that DirSkills catalogs from Xiangyue-Zhang/auto-deep-researcher-24x7, the repository it ships in. Its siblings there include Conf Search, Daily Papers and Experiment Status. Each one is a separate skill with its own page in this directory, installs the same way Auto Experiment does, and is maintained by Xiangyue-Zhang in that same repository. The rest of the collection is listed on the Xiangyue-Zhang/auto-deep-researcher-24x7 page.
How does Auto Experiment compare to other Automation skills?
Auto Experiment ranks #689 by stars among the 1523 Automation skills in this catalog. The most-starred ones next to it are Autonomous Loops, Autonomous Agent Harness and Automation Audit Ops. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Auto Experiment against them. Open each page to compare what they document and how they install.