Tao Hu
Community@dongzhuoyao · Munich
Agent Skills by Tao Hu
Showing 82 vetted skills indexed across 2 GitHub repositories.
dongzhuoyaolumi-inode-status
Check LUMI scratch and home directory inode quota usage and return a go/no-go launch verdict.
dongzhuoyaosue-parallelism-discussion
Validate ML/HPC parallelism plans for GPU-occupancy-first node packing.
dongzhuoyaosue-sandbox-quota-calculation
Convert HPC quota units and validate Slurm job consumption across LUMI and Snellius.
dongzhuoyaosue-cronjob
Manage SUE-tagged cron entries across Unix, Codex, Kimi, and RunPod backends.
dongzhuoyaosue-interface-check
Audit ML experiment codebases for portable scale-up readiness across HPC and cloud backends.
dongzhuoyaosue-cleanup
Archive verified bulk output directories into tar archives to free inodes.
dongzhuoyaosue-tdd-demo-cifar10
Run a full SUE pipeline rehearsal on a DiT CIFAR-10 workspace and validate path rules.
dongzhuoyaokimi-code-review
Review code diffs with the Kimi Code CLI and relay severity-marked findings.
dongzhuoyaosue-cluster-info
Query live Slurm job limits and GPU hardware specs for LUMI and Snellius clusters.
dongzhuoyaosingle-gpu-dryrun
Run small single-GPU probe jobs to validate ML experiments before production.
dongzhuoyaosue-run-preparation
Prepare ML/HPC scale-up experiment preflight runs across sandbox backends.
dongzhuoyaosue-nm5-slurm-status
Retrieve read-only NM5 Slurm scheduler state, job status, and GPU-hour accounting data.
dongzhuoyaoexperiment-env-installer
Install reproducible Python environments for ML experiments with uv across compute backends.
dongzhuoyaosue-synccode
Synchronize DeepResearch experiment code to remote HPC and cloud sandboxes.
dongzhuoyaosue-doc-code-consistency
Audit DeepResearch SUE framework documentation for consistency and doc-code alignment.
dongzhuoyaocodebase-inode-audit
Audit ML experiment codebases for inode-heavy file patterns before scale-up runs.
dongzhuoyaosue-overall-audit
Audit SUE skill family definitions for contract drift and cross-tool compatibility.
dongzhuoyaowandb-result-analysis
Validate Weights & Biases experiment metrics against local CSV/JSONL ledgers.
dongzhuoyaosue-reset
Archive epoch artifacts and reset ML/HPC experiment workspaces.
dongzhuoyaosue-exp-naming
Generate and validate SUE scale-up experiment identifiers for W&B, Slurm, and result ledgers.
dongzhuoyaosue-update-lesson
Record durable lessons learned from SUE scale-up workflows into persistent memory files.
dongzhuoyaosue-doc-demo-page
Synchronize SUE demo workspaces with documentation pages, navbar entries, and generator configurations.
dongzhuoyaosue-dataset-preparation
Prepare ML/HPC datasets with inode-safe formats and storage compliance checks.
dongzhuoyaosue-doc-update
Regenerate SUE framework HTML docs and GitHub Pages from canonical Markdown sources.