remote-gpu-trainer SKILL.md skillA

remote-gpu-trainer from hanyuyuan6/remote-gpu-trainer is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add hanyuyuan6/remote-gpu-trainer/remote-gpu-trainer@git:20261108.3117ed4

Pins git:20261108.3117ed4 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
hanyuyuan6/remote-gpu-trainer
Path
SKILL.md
Versions recorded
11
Latest version
git:20261108.3117ed4, committed Nov 8, 2026
First version recorded
Aug 2, 2026
Size
13,725 bytes
Repository stars
42
License
MIT

AutoDL/vast.ai/RunPod/Lambda/Paperspace远程GPU启动、调试、监控、续训、并行、停止及卡住/OOM/NaN/全零指标排查;可验证状态/selection-best checkpoint/拉回结果。释放实例销毁机器及数据盘,不等同关机。长期镜像/恢复→mirror-research-artifacts。.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Remote GPU Trainer

## Mission

Move one experiment through **RUN → VERIFY → DELIVER** with the least control work that safely advances the
user's outcome. This is the compute/control layer and owns one compute attempt. When the task spans several
runs, nodes, independent side-effect lanes, or paper-wide closeout, route scheduling and state projection to
`research-artifact-hygiene` (closeout ledger); do not duplicate a second controller here. Companion skills named
anywhere in this skill (`research-artifact-hygiene`, `mirror-research-artifacts`, `superpowers:*`,
`huggingface-skills:*`) are optional separate installs: without one, do the equivalent step with your own tools.

## Start with the outcome, not a package

Before tools, write a compact internal decision:

1. desired result and current verified state;
2. exact next executable action;
3. consequence level below;
4. stop condition and where the result will live.

Do not preload incident catalogues, every platform profile, or a whole project SSoT. Pick one profile and one
phase reference. Keep large evidence out of model context: verify identity/schema, parse task-relevant fields,
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add hanyuyuan6/remote-gpu-trainer/remote-gpu-trainer@sha256:f372e6b911103526

The line at the top of this page pins the label git:20261108.3117ed4, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_xzvapsf6y2hpza6y.svg)](https://markdownregistry.com/a/art_xzvapsf6y2hpza6y)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 11 versions of this file, between Aug 2, 2026 and Nov 8, 2026, about one every 10 days.

versioncommittedcommitsizeaudit
git:20261108.3117ed4 latest2026-11-08 3117ed4 13,725 BA view · diff
git:20261028.fe01c552026-10-28 fe01c55 13,453 BA view · diff
git:20261022.566669c2026-10-22 566669c 12,903 BA view · diff
git:20261013.71ddabd2026-10-13 71ddabd 12,812 BA view · diff
git:20261009.9cd83972026-10-09 9cd8397 12,707 BA view · diff
git:20260917.34be1262026-09-17 34be126 12,474 BA view · diff
git:20260915.96b55122026-09-15 96b5512 12,568 BA view · diff
git:20260915.e44ada82026-09-15 e44ada8 12,369 BA view · diff
git:20260810.892e5672026-08-10 892e567 22,294 BA view · diff
git:20260810.974a39f2026-08-10 974a39f 21,970 BA view · diff
git:20260802.37ed2082026-08-02 37ed208 19,043 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (13725 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

hanyuyuan6/remote-gpu-trainer · 42 stars · license MIT · pushed 2026-11-08 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_xzvapsf6y2hpza6y
GET https://markdownregistry.com/api/v1/resolve?ref=hanyuyuan6/remote-gpu-trainer/remote-gpu-trainer
GET https://markdownregistry.com/api/v1/blob/f372e6b91110352669dce96fd155ce81235beb0fb253466785419b3807c3821e

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

Browse by kind, by grade A, or by owner.