AGENTS.md@src/srtctl/benchmarksA
AGENTS.md@src/srtctl/benchmarks from nvidia/srt-slurm is an AGENTS.md file: the standing instructions coding agents read in that repo.
mdr add nvidia/srt-slurm/src/srtctl/benchmarks/AGENTS.md@git:20261003.7f76be2Pins git:20261003.7f76be2 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 14 of 14 checks passed
- Kind
- AGENTS.md file
- Repository
- nvidia/srt-slurm
- Path
src/srtctl/benchmarks/AGENTS.md- Versions recorded
- 2
- Latest version
- git:20261003.7f76be2, committed Oct 3, 2026
- First version recorded
- Sep 29, 2026
- Size
- 1,482 bytes
- Repository stars
- 93
- License
- a license file GitHub does not identify
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
# Benchmarks
Runners live here; the shell scripts they launch live under `benchmarks/scripts/` and are excluded from Ruff and `ty`.
## Adding a New Benchmark
1. Create `benchmarks/<name>.py` with a class decorated `@register_benchmark("<type>")` (`benchmarks/base.py`). Subclass `BenchmarkRunner`, or `AIPerfBenchmarkRunner` for an AIPerf-driven benchmark.
2. Implement the abstract members: `name` and `script_path` (properties; the script is under `SCRIPTS_DIR`), `validate_config(config)` returning a list of error strings, and `build_command(...)`. Override `get_container_image`, `get_container_mounts`, or `get_environment` only when the benchmark needs something other than the defaults.
3. Declare the `BenchmarkConfig` fields the runner reads beyond `SHARED_BENCHMARK_FIELDS` in `config_fields` (`benchmark_config_fields()` combines the two). A field set for a type whose runner does not read it is rejected under schema 2 and warned about under schema 1 (`SrtConfig` validation in `core/schema.py`).
4. Add the script under `benchmarks/scripts/<name>/`.
5. Import the module from `benchmarks/__init__.py` so registration runs.
…Read the whole file at its exact version.
How to install
mdr add nvidia/srt-slurm/src/srtctl/benchmarks/AGENTS.md@sha256:e0810b07130850deThe line at the top of this page pins the label git:20261003.7f76be2, 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_b7lvgpqpdsefslez)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 2 versions of this file, between Sep 29, 2026 and Oct 3, 2026, about one every 5 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261003.7f76be2 latest | 2026-10-03 | 7f76be2 | 1,482 B | A | view · diff |
| git:20260929.92fbe04 | 2026-09-29 | 92fbe04 | 1,488 B | A | view |
Audit of the latest version
- pass: Size between 200 bytes and 200 KB (1482 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 · 93 stars · license a license file GitHub does not identify · pushed 2026-10-20 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_b7lvgpqpdsefslez GET https://markdownregistry.com/api/v1/resolve?ref=nvidia/srt-slurm/src/srtctl/benchmarks/AGENTS.md GET https://markdownregistry.com/api/v1/blob/e0810b07130850de68b64de3956f5af37a76103856ec7cd0a9729ed18c66e8f1
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: how to write a good AGENTS.md · AGENTS.md vs CLAUDE.md · which agent file does what.