Home / lu1sdv / skillsmd · vast-experiment-ops/SKILL.md · GitHub

vast-experiment-ops skillA

vast-experiment-ops is agent-read markdown (skill) from lu1sdv/skillsmd: Use when running or reporting ML experiments on rented Vast.ai GPU instances — diagnosing strict FP32 equivalence checks that pass on CPU but fail on CUDA (cuDNN TF32), parsing JSON metrics collected over SSH, and keeping held-out/test-scope reporting accurate..

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# vast-experiment-ops

Error-informed operational playbook for renting, running, and reporting ML
experiments on Vast.ai. It supplements the platform docs and the instance's own
guide; it replaces neither.

## Quick reference

| Situation | Action |
|---|---|
| Strict FP32 equivalence passes on CPU, fails on CUDA | Re-run the unchanged check in a fresh process with `NVIDIA_TF32_OVERRIDE=0` set before Python starts |
| `JSONDecodeError` parsing remote metrics | Capture stdout/stderr separately, parse stdout only; collector emits exactly one JSON document on stdout |
| Asked whether the test split was used | Answer from this run's queue config and result metadata plus prior artifacts — never open test data to answer |
| Reporting progress | Include UTC time, snapshot age, active trial/epoch, and complete/pruned/failed/running counts |
| Sizing a job | Check cgroup CPU/RAM limits, free GPU memory, free disk, and storage persistence — host totals mislead |

## Ground rules

- Get the instance address and experiment paths from the user or project
  config. Never reuse an address remembered from this skill.
- Read the instance's own guide (e.g. `/etc/vast-agents-guide.md`) before
…

Read the whole file at its exact version.

How to install

Latest version
mdr add lu1sdv/skillsmd/vast-experiment-ops@git:20260919.01355af
Exact content
mdr add lu1sdv/skillsmd/vast-experiment-ops@sha256:8166e7a0e1233159

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_ky425osxu6b26s6l.svg)](https://markdownregistry.com/a/art_ky425osxu6b26s6l)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260919.01355af latest2026-09-19 01355af 6,875 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, 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 (6875 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

lu1sdv/skillsmd · 2 stars · license none · pushed 2026-09-22 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_ky425osxu6b26s6l
GET https://markdownregistry.com/api/v1/resolve?ref=lu1sdv/skillsmd/vast-experiment-ops
GET https://markdownregistry.com/api/v1/blob/8166e7a0e1233159c1a70cf57686305663ac26dd341e25786183f4e3b1bf3e36

Your agent 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.

More from lu1sdv/skillsmd

bug-bounty-reporting skill
lu1sdv/skillsmd · bug-bounty-reporting/SKILL.md · A skill for creating a report in a bug bounty vulnerability assessment.Use when an agent drafts, revises, freezes, or…
git:20260922.6749075 · audit A · 2 stars
bugbounty-poc-video-runner skill
lu1sdv/skillsmd · bugbounty-poc-video-runner/SKILL.md · Designs a one-command, triager-friendly live security PoC video runner with staged evidence, expected-versus-actual…
git:20260829.50b9f72 · audit A · 2 stars
codeql skill
lu1sdv/skillsmd · codeql/SKILL.md · Scans a codebase for security vulnerabilities using CodeQL's interprocedural data flow and taint tracking analysis…
git:20260516.f0a764e · audit A · 2 stars
cook skill
lu1sdv/skillsmd · cook/SKILL.md · Use when user says "cook", "let it cook", wants iterative refinement with quality gates, wants to race multiple…
v1.0.0 · audit A · 2 stars
ctfd skill
lu1sdv/skillsmd · ctfd/SKILL.md · Use when developing, debugging, deploying, or maintaining any CTFd instance or fork - Docker stack issues, plugin…
git:20260314.993db98 · audit A · 2 stars
debate skill
lu1sdv/skillsmd · debate/SKILL.md · Use when a technical decision has 2–4 nameable positions and the user wants a recommendation, not open-ended…
git:20260419.6f6fafc · audit A · 2 stars
feature-engineering skill
lu1sdv/skillsmd · feature-engineering/SKILL.md · Use when building or improving time series forecasting models and the user asks about exogenous variables, calendar…
git:20260411.b88fc6f · audit A · 2 stars
fuzzgpt skill
lu1sdv/skillsmd · fuzzgpt/SKILL.md · Use when you want to fuzz any software target — library, compiler, interpreter, parser, kernel/syscall, DB engine, SMT…
git:20260922.ae9c57d · audit A · 2 stars
glitchtip skill
lu1sdv/skillsmd · glitchtip/SKILL.md · Use when deploying, configuring, integrating, or troubleshooting GlitchTip — including self-hosted installation, SDK…
git:20260314.993db98 · audit A · 2 stars
hermes-mr-watch skill
lu1sdv/skillsmd · hermes-mr-watch/SKILL.md · Use when adding GitLab MRs to the Hermes watcher on ssh-vhs, editing or redeploying watch_gitlab_mr_findings.py…
git:20260922.2b9d11c · audit A · 2 stars
hermes-self-evolution skill
lu1sdv/skillsmd · hermes-self-evolution/SKILL.md · Replicate the Hermes Agent self-evolving loop for agents that should learn from experience: persistent declarative…
git:20260424.3addf1c · audit A · 2 stars
jev skill
lu1sdv/skillsmd · jev/SKILL.md · Use when writing or improving Jev programs — the TypeSafe judgment model behind POST /v1/systemone. Covers primitive…
git:20260922.b382a88 · audit A · 2 stars

Every file in lu1sdv/skillsmd

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