菜单
How to answer user questions about what models are available, filter by domain/task/license, and compare models in AI4Science Studio.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Applies to Earth-system ML in AI4Science Studio—climate, weather, geospatial data, and model folders under earth_science/models/. Includes validated AMD/ROCm HPC patterns for earth science models.
Applies to material science and chemistry ML recipes under material_science/models/ in AI4Science Studio.
Runs the AMD AI agents bottleneck-analysis workflow on a HydraGNN or ORBIT-2 training job using TraceLens + Omnistat user-mode + paired analyst/verifier subagents, or drives the iterative perf-optimizer-loop (accept/revert on throughput or epoch time). Use when the user wants to analyze a perf run, analyze a 2-node HydraGNN run, find bottlenecks with TraceLens/Omnistat, run the AMD AI agents on a job, run the perf-optimizer-loop, tune ORBIT-2 throughput, or optimize HydraGNN training performance.
Applies when working in the AI4Science Studio repository. Describes domain layout, model slug rules, where recipes live, safety expectations, and AMD/ROCm HPC patterns (Apptainer and Docker) validated in practice.
Step-by-step instructions for how Cursor should run any model in AI4Science Studio — read model.yaml, check preflight, select container runtime, launch.
Applies to healthcare and life sciences ML recipes under healthcare/models/ in AI4Science Studio, with privacy and non-clinical-use guardrails.
| name | ai4science-discover |
| description | How to answer user questions about what models are available, filter by domain/task/license, and compare models in AI4Science Studio. |
Use this skill when a user asks what models are available, wants to compare models, or asks questions like "what can I do with this repo?"
Read models.yaml at the repo root. It lists every model with:
slug — folder namedomain — earth_science, material_science, healthcare, physics_simulation, protein_foldinghf_id — Hugging Face model id (or N/A)license — SPDX idtask — one-line descriptiontasks_available — list of recipe tasks (inference, train, finetune, ensemble)path — relative path to model folderParse models.yaml and present a table:
| Model | Domain | Task | Available recipes |
|---|---|---|---|
| StormCast | Earth science | Weather prediction | inference, ensemble |
| ... | ... | ... | ... |
When the user asks for a subset:
domain fieldtasks_available (e.g. "which models support fine-tuning?" → filter for finetune in tasks_available)license (e.g. "which models are MIT licensed?")hf_id != N/A for models with HF weightsWhen the user wants details on a specific model:
<path>/model.yaml for structured metadata<path>/README.md for prose description<path>/examples/README.md for available scripts<path>/recipes/ subdirectories for available recipesWhen the user asks to compare two or more models:
model.yamlWhen asked "which models are ready to run?":
models.yaml, check:
examples/docker_run.sh exist?examples/preflight_*.py exist?recipes/*/README.md exist?examples/sbatch_*_amd.sh exist?