slurm-job-sizing SKILL.md skillA
slurm-job-sizing from nvidia/srt-slurm is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add nvidia/srt-slurm/slurm-job-sizing@git:20261002.ed509b8Pins git:20261002.ed509b8 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- nvidia/srt-slurm
- Path
src/srtctl/dsight/skills/slurm-job-sizing/SKILL.md- Versions recorded
- 1
- Latest version
- git:20261002.ed509b8, committed Oct 2, 2026
- Size
- 2,648 bytes
- Repository stars
- 95
- License
- a license file GitHub does not identify
Size Slurm allocations, warmup and traffic for short debugging jobs or full srt-slurm benchmarks.
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
# Size Slurm jobs to the task Use **non-preemptible resources** and preserve the required GPU count/type, topology and experiment controls. Requested walltime is the main adjustable scheduling lever: shorter jobs can fit scheduling gaps, though numeric priority depends on the site. - **Define the evidence needed.** Reuse existing captures or local/CPU checks when sufficient; otherwise choose the smallest experiment that answers the question. - **Debug, build, test or verify:** prefer short, independent hypothesis tests **in parallel**, within authorized aggregate resources. Target **at most 30 minutes per whole job**. Reduce warmup and measured traffic to the minimum that exercises the relevant path; preserve the concurrency and request shape needed to reproduce it. Isolate each job's state and outputs; serialize actual dependencies. Split preparation first if the budget does not fit, and justify unavoidable exceptions. Label reduced runs as diagnostic. - **Full load point:** search history before choosing walltime. Find multiple comparable runs by model, hardware, engine, workload, concurrency, cache state and instrumentation. Inspect their …
Read the whole file at its exact version.
How to install
mdr add nvidia/srt-slurm/slurm-job-sizing@sha256:d7753fd777912249The line at the top of this page pins the label git:20261002.ed509b8, 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.
[](https://markdownregistry.com/a/art_amrrjm6w5nluwy5h)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 2, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (2648 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
nvidia/srt-slurm · 95 stars · license a license file GitHub does not identify · pushed 2026-11-06 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_amrrjm6w5nluwy5h GET https://markdownregistry.com/api/v1/resolve?ref=nvidia/srt-slurm/slurm-job-sizing GET https://markdownregistry.com/api/v1/blob/d7753fd777912249abf50391def8f8c348d5039e990037e99f5fb6cff9aad694
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.