{
  "name": "remote-gpu-trainer",
  "namespace": "@sickn33",
  "version": "1.0.0",
  "description": "Deploy, monitor, and debug long GPU jobs on RENTED/remote instances (AutoDL, RunPod, vast.ai, Lambda, Slurm, K8s): teardown/billing safety, spot resilience, resumable checkpointing, OOM/NaN triage.",
  "category": "ai-ml",
  "tags": [
    "ai-ml"
  ],
  "files": [
    ".gitattributes",
    ".gitignore",
    "LICENSE",
    "README.md",
    "SKILL.md",
    "evals/README.md",
    "evals/RESULTS.md",
    "evals/cases.jsonl",
    "evals/run_evals.py",
    "examples/autodl_sweep/README.md",
    "examples/autodl_sweep/queue_1.txt",
    "profiles/_schema.md",
    "profiles/autodl.md",
    "profiles/china.md",
    "profiles/generic-ssh.md",
    "profiles/lambda.md",
    "profiles/paperspace.md",
    "profiles/runpod.md",
    "profiles/vastai.md",
    "references/china-network.md",
    "references/gotchas_universal.md",
    "references/lifecycle_checklist.md",
    "references/monitoring_patterns.md",
    "references/multinode.md",
    "references/parallel_ablation.md",
    "references/principles.md",
    "references/self-improvement.md",
    "references/spot-resilience.md",
    "references/ssh_transport.md",
    "references/training/by-domain.md",
    "references/training/checkpoint-resume.md",
    "references/training/convergence-debugging.md",
    "references/training/data-pipeline.md",
    "references/training/distributed-launch.md",
    "references/training/oom-memory.md",
    "references/training/precision-stability.md",
    "references/training/throughput-profiling.md",
    "scripts/aggregate_to_fs.sh",
    "scripts/check_staleness.py",
    "scripts/download_loop.sh",
    "scripts/gpu_health.sh",
    "scripts/health_patrol.sh.template",
    "scripts/mem_monitor.sh",
    "scripts/reap_vram_zombies.sh",
    "scripts/run_one.sh.template",
    "scripts/run_queue.sh.template",
    "scripts/setup-china-mirrors.sh",
    "scripts/verify_local.py"
  ],
  "kaged_version": ">=0.4.0",
  "license": "MIT"
}