← All skills
run-experiment
80 / 100 · Low risk
Fine to try.
By wanshuiyin/Auto-claude-code-research-in-sleep (GitHub) · research · 19,020 stars · scanned at commit 8f49392
Deploy and run ML experiments on local, remote, Vast.ai, or Modal serverless GPU. Use when user says "run experiment", "deploy to server", "跑实验", or needs to launch training jobs.
Works with any agent that supports SKILL.md skills.
What this skill can do
Runs commands on your computer
The author does not mention this.
Found in 18 places.
Show 15 more
- skills/run-experiment/SKILL.md, line 85
rsync -avz --prune-empty-dirs \ - skills/run-experiment/SKILL.md, line 107
git add -A && git commit -m "sync: experiment deployment" && git push - skills/run-experiment/SKILL.md, line 119
rsync -avz -e "ssh -p <PORT>" \ - skills/run-experiment/SKILL.md, line 131
ssh -p <PORT> root@<HOST> "pip install -q torch==<pinned>" # phase 1: pins - skills/run-experiment/SKILL.md, line 137
scp -P <PORT> requirements.txt root@<HOST>:/workspace/ - skills/run-experiment/SKILL.md, line 174
ssh <server> "wandb status" # should show logged in - skills/run-experiment/SKILL.md, line 187
ssh <server> "screen -dmS <exp_name> bash -c '\ - skills/run-experiment/SKILL.md, line 197
ssh -p <PORT> root@<HOST> "screen -dmS <exp_name> bash -c '\ - skills/run-experiment/SKILL.md, line 224
CUDA_VISIBLE_DEVICES=<gpu_id> python <script> <args> 2>&1 | tee <log_file> - skills/run-experiment/SKILL.md, line 236
ssh <server> "screen -ls" - skills/run-experiment/SKILL.md, line 241
ssh -p <PORT> root@<HOST> "screen -ls" - skills/run-experiment/SKILL.md, line 246
modal app list # Check app is running - skills/run-experiment/SKILL.md, line 267
rsync -avz -e "ssh -p <PORT>" root@<HOST>:/workspace/project/results/ ./results/ - skills/run-experiment/SKILL.md, line 272
scp -P <PORT> root@<HOST>:/workspace/*.log ./logs/ - skills/run-experiment/SKILL.md, line 277
vastai destroy instance <INSTANCE_ID>
Downloads and installs other software
The author does not mention this.
Found in 6 places.
Show 3 more
Other copies in this repository
wanshuiyin/Auto-claude-code-research-in-sleep has 1 more folder with a skill named run-experiment. This page shows the lowest-scoring copy.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'run-experiment' ] || [ -L ~/.agents/skills/'run-experiment' ]; then echo 'A skill folder named run-experiment already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep' "$d" && git -C "$d" checkout --quiet '8f493923f28c3f991dcca1287db97aa780d0969b' -- 'skills/run-experiment' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/run-experiment' ~/.agents/skills/'run-experiment' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'run-experiment' ~/.claude/skills/'run-experiment' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
The badge for wanshuiyin/Auto-claude-code-research-in-sleep shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/wanshuiyin/Auto-claude-code-research-in-sleep/)
Score history
| Date | Version | Score |
|---|
| 2026-11-11 | 8f49392 | 80 |
| 2026-11-08 | 2901234 | 80 |
| 2026-11-07 | 2901234 | 77 |