Borrowing it
Nothing to install: this file belongs to AMDResearch/ai4science-studio. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/AMDResearch/ai4science-studio/main/.cursor/skills/ai4science-run-models/SKILL.mdgit clone --depth 1 https://github.com/AMDResearch/ai4science-studioWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/amdresearch/ai4science-studio/ai4science-run-models)<a href="https://agentmods.dev/skills/amdresearch/ai4science-studio/ai4science-run-models"><img src="https://agentmods.dev/badge/skills/amdresearch/ai4science-studio/ai4science-run-models/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/amdresearch/ai4science-studio/ai4science-run-models"><img src="https://agentmods.dev/badge/skills/amdresearch/ai4science-studio/ai4science-run-models.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00038 | $0.01393 |
| Opus 5 | $0.00019 | $0.00696 |
| Sonnet 5 | $0.00008 | $0.00279 |
| Haiku 4.5 | $0.00004 | $0.00139 |
Grade A, and why
ai4science-run-models scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Running models in AI4Science Studio
Use this skill when a user asks to run, execute, or launch any model in the repository.
Step 0: Check for cluster config
Before anything else, check if a cluster config exists:
# Check both locations
test -f .cluster-config.yaml && echo "repo-local" || \
test -f ~/.config/ai4science-studio/cluster.yaml && echo "user-level" || \
echo "missing"
If missing, tell the user: "No cluster configuration found. Running /init-cluster to detect your cluster environment." Then run the init-cluster flow (auto-discover GPU, SLURM, containers, paths) before proceeding.
If a config exists, read it and use its values as defaults for SLURM partition, account, container runtime, GPU arch, and scratch paths throughout the run. Still confirm with the user if any value is empty or looks stale.
Step 1: Identify the model
- Read
models.yamlat the repo root to find the model by name, slug, or HF id. - Read
<domain>/models/<slug>/model.yamlfor full metadata: recipes, env vars, container image, hardware requirements.
Step 2: Determine the task
Match the user's request to a recipe in model.yaml:
- "run inference" / "predict" / "generate" → look for
task: inference - "train" / "fine-tune" / "pair-tune" → look for
task: trainortask: finetune - "ensemble" → look for
task: ensemble
Step 3: Ask the user for required inputs
Read the env_vars section of model.yaml. For every variable marked required: true that has default: null, ask the user for the value. Present the question with the variable's description.
For common patterns, always offer three options — provide manually, generate/build, or auto-discover:
- SIF path → "Do you have an Apptainer SIF file? (Yes / No — I'll generate the pull command / Auto-discover — I'll search the filesystem)"
- Overlay → "Do you have a pre-built overlay? (Yes / No, build one / No, skip overlay / Auto-discover — I'll search for an existing one)"
- Upstream repo → "Do you have the repo cloned? (Yes / No — I'll generate the clone command / Auto-discover — I'll search the filesystem)"
- SLURM partition/account → "How should I determine your partition/account? (Provide manually / Auto-discover — I'll query SLURM)"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 114 lines · 38 tokens per session scan A 4365d2c38bfc
ai4science-run-models is a skill published in the GitHub repository AMDResearch/ai4science-studio (4 stars, last pushed 1mo ago), licensed MIT. It adds 38 tokens to every session and 1,393 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
annotating-variants
Annotates VCF variants and normalizes HGVS nomenclature with public, license-free annotators (Ensembl VEP REST, VEP/SnpEff/ANNOVAR offline) and links variants to gnomAD population frequencies and the clinical context OpenMed extracts. Use when the user wants to predict variant consequences, map HGVS to genomic…
batch-processing-clinical-text
Run large-scale batch NER, PII extraction, or de-identification over many clinical notes on-device with OpenMed, with sharding, checkpointing, resumability, and append-only JSONL output. Use when the user needs to process a corpus or folder of notes, de-identify a dataset, run NER over thousands of documents, build a…
coding-hcc-risk-adjustment
Maps chronic conditions extracted by OpenMed to CMS-HCC V28 risk-adjustment categories and estimates a RAF (Risk Adjustment Factor) score as decision support. Use when the user wants to surface risk-adjustable diagnoses from notes, map ICD-10-CM codes to HCC categories, estimate or reconcile a patient/panel RAF, find…
mapping-to-snomed
Maps clinical concept spans extracted by OpenMed to SNOMED CT concepts through a USER-SUPPLIED terminology server (the user's own Ontoserver, Snowstorm, or UMLS/UTS), never a bundled vocabulary. Use when the user wants to code findings, disorders, procedures, body structures, or substances to SNOMED CT, run an ECL…
structuring-radiology-reports
Converts free-text radiology narratives into structured findings and impression — with measurements, laterality, anatomy, and follow-up recommendations — after OpenMed NER. Use when the user has a CT/MRI/X-ray/ultrasound/mammography report and needs the sections split (technique, comparison, findings, impression)…
coding-icd10
Suggests candidate ICD-10-CM diagnosis codes (and ICD-10-PCS procedure codes) for diagnoses and procedures extracted by OpenMed, with rationale and a human-coder caveat. Use when the user wants to code a problem list, map a diagnosis span to a billable ICD-10-CM code, route a finding to the right chapter, cross-walk…