Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention
- Stars
- 14,234
- Version
- v1
- Copies
- 0
Overview
Waxell's checks found 8 findings to review, highest severity high.
- Highest severity
- high
- Findings
- 8
- What ran
- Static analysis
A clean result means nothing alarming in what is published, not a guarantee. How the checks work
We don't republish the skill's instructions. Read them on GitHub at the reviewed commit.
Security
Security scan
Checked 2 days ago
Result
Review findings Waxell's checks found 8 findings to review, highest severity high.
Checked with static analysis: servers it launches, packages it installs at runtime, scripts piped to a shell and bundled binaries. Nothing was run. How the checks work
Findings (8)
high · 8
- high
Tells Claude to install unpinned packages (pre-commit, Next, the, pre-commit, hooks, must …)
references/other.md:1099 · skill_risk:S2
- high
Tells Claude to install unpinned packages (pytest, Simply, invoke, pytest, to, run …)
references/other.md:1107 · skill_risk:S2
- high
Tells Claude to install unpinned packages (datasets)
references/tutorials.md:727 · skill_risk:S2
- high
Tells Claude to install unpinned packages (deepspeed, see, more, details)
references/tutorials.md:5030 · skill_risk:S2
- high
Tells Claude to install unpinned packages (deepspeed)
references/tutorials.md:6126 · skill_risk:S2
- high
Tells Claude to install unpinned packages (torch)
references/tutorials.md:6467 · skill_risk:S2
- high
Tells Claude to install unpinned packages (oneccl_bind_pt, https://developer.intel.com/ipex-whl-stable-cpu)
references/tutorials.md:6485 · skill_risk:S2
- high
Tells Claude to install unpinned packages (torch, torch_npu)
references/tutorials.md:6515 · skill_risk:S2
Versions
Version history
Each distinct content we have seen, newest first, with the verdict it got at the time.
One version so far: v1 (1.0.0), seen 2026-11-03.
See and control which skills your people use
Inside your organization, Waxell scans each skill and plugin before anyone installs it, approves what passes and publishes it to a private marketplace, and puts every MCP tool call behind one governed URL.
30 minutes. No commitment.
SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF
More research skills and plugins
training-llms-megatronorchestra-researchTrains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training… Review findingsResearchSales & marketing14k
pytorch-fsdp2orchestra-researchAdds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed… No issuesResearchDocuments14k
pytorch-lightningorchestra-researchHigh-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal… Review findingsResearchEngineering14k
ray-trainorchestra-researchDistributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in… Review findingsResearchAI & agents14k