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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/brevdev/workshop-build-an-agent --skill code-review命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Operator/host side of running the NVIDIA "Build an Agent" DevX workshop inside an OpenShell/NemoClaw sandbox. ⚠️ NOT the default workshop-setup path: for a generic "set up the workshop" request, use the `setup-workshop` skill (the standard bare-metal/Brev/AI-Workbench installer). Invoke THIS skill ONLY if the user names it explicitly, or explicitly says the workshop should run with "NemoClaw" or inside an "OpenShell" sandbox. When it does apply, use it when YOU are on the sandbox HOST (outside the sandbox — the machine running the OpenShell gateway + docker, e.g. via Claude Code) and the user wants the workshop set up in a sandbox: stage/apply the egress policy (PyPI, NIM /v1/ranking, scoped GitHub clone), optionally stage the NVIDIA API key, kick the in-sandbox agent (which runs the `setup-workshop-nemoclaw` skill), then open the inbound path (openshell forward service + SSH/Teleport port-forward) so the user can open the JupyterLab token URL. Also covers sandbox lifecycle pitfalls (container restarts are on
Set up the NVIDIA "Build an Agent" DevX workshop as a working JupyterLab environment from INSIDE a locked-down OpenShell/NemoClaw sandbox, and hand the user the token URL + access commands. Use this when the user asks to set up / run / access the Build-an-Agent workshop and YOU are the agent running inside the sandbox (repo at /sandbox/workshop-build-an-agent). NOT the generic `setup-workshop` skill (a bare-metal GPU-host installer that needs sudo/Docker/CUDA and cannot run here), and NOT the host side — operators outside the sandbox use `setup-workshop-nemoclaw-operator`. This is the sandbox-native path: uv venv + pinned CPU deps, an LD_PRELOAD netlink shim to get Jupyter kernels past the seccomp AF_NETLINK block, a hand-built labextension bridge, launcher path rewrites, %pip-cell neutralization, and single-server discipline. Modules 1-3 (CPU) work end-to-end; module 4's training notebooks (GPU: torch/unsloth) do not run here by design, and module 7's optional cudf exercise falls back to CPU.
Operator/host side of running the NVIDIA "Build an Agent" DevX workshop inside an OpenShell/NemoClaw sandbox. ⚠️ NOT the default workshop-setup path: for a generic "set up the workshop" request, use the `setup-workshop` skill (the standard bare-metal/Brev/AI-Workbench installer). Invoke THIS skill ONLY if the user names it explicitly, or explicitly says the workshop should run with "NemoClaw" or inside an "OpenShell" sandbox. When it does apply, use it when YOU are on the sandbox HOST (outside the sandbox — the machine running the OpenShell gateway + docker, e.g. via Claude Code) and the user wants the workshop set up in a sandbox: stage/apply the egress policy (PyPI, NIM /v1/ranking, scoped GitHub clone), optionally stage the NVIDIA API key, kick the in-sandbox agent (which runs the `setup-workshop-nemoclaw` skill), then open the inbound path (openshell forward service + SSH/Teleport port-forward) so the user can open the JupyterLab token URL. Also covers sandbox lifecycle pitfalls (container restarts are on
基于 SOC 职业分类
正在显示 SKILL.md
| name | code_review |
| description | Systematic approach to reviewing code for quality and correctness |
You are now operating as a code reviewer. Follow this systematic approach.
Structure your review as:
## Summary
[One sentence overall assessment]
## What's Good
- [Positive point 1]
- [Positive point 2]
## Suggestions
1. [Issue]: [Specific recommendation]
2. [Issue]: [Specific recommendation]
## Priority
[High/Medium/Low] - [Reason]