This skill should be used when the user wants to set up, install, deploy, bootstrap, or "spin up" the Build-an-Agent workshop (a.k.a.
$ npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install brevdev/workshop-build-an-agent setup-workshop --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/brevdev/workshop-build-an-agent.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/setup-workshop .claude/skills/setup-workshop && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "setup-workshop" agent skill from https://github.com/brevdev/workshop-build-an-agent/tree/main/.agents/skills/setup-workshop into .claude/skills/setup-workshop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-workshop", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/brevdev/workshop-build-an-agent/tree/main/.agents/skills/setup-workshopType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install brevdev/workshop-build-an-agent setup-workshop --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/brevdev/workshop-build-an-agent.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/setup-workshop .agents/skills/setup-workshop && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "setup-workshop" agent skill from https://github.com/brevdev/workshop-build-an-agent/tree/main/.agents/skills/setup-workshop into .agents/skills/setup-workshop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-workshop", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install brevdev/workshop-build-an-agent setup-workshop --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/brevdev/workshop-build-an-agent.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/setup-workshop .cursor/skills/setup-workshop && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "setup-workshop" agent skill from https://github.com/brevdev/workshop-build-an-agent/tree/main/.agents/skills/setup-workshop into .cursor/skills/setup-workshop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-workshop", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/brevdev/workshop-build-an-agent.git --path .agents/skills/setup-workshop--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install brevdev/workshop-build-an-agent setup-workshop --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/brevdev/workshop-build-an-agent.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/setup-workshop .gemini/skills/setup-workshop && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "setup-workshop" agent skill from https://github.com/brevdev/workshop-build-an-agent/tree/main/.agents/skills/setup-workshop into .gemini/skills/setup-workshop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-workshop", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install brevdev/workshop-build-an-agent setup-workshopInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/brevdev/workshop-build-an-agent.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/setup-workshop .github/skills/setup-workshop && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "setup-workshop" agent skill from https://github.com/brevdev/workshop-build-an-agent/tree/main/.agents/skills/setup-workshop into .github/skills/setup-workshop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-workshop", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install brevdev/workshop-build-an-agent setup-workshop --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/brevdev/workshop-build-an-agent.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/setup-workshop .opencode/skills/setup-workshop && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "setup-workshop" agent skill from https://github.com/brevdev/workshop-build-an-agent/tree/main/.agents/skills/setup-workshop into .opencode/skills/setup-workshop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-workshop", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
setup-workshopThis skill should be used when the user wants to set up, install, deploy, bootstrap, or "spin up" the Build-an-Agent workshop (a.k.a.
Setup Workshop is an agent skill from brevdev/workshop-build-an-agent. This skill should be used when the user wants to set up, install, deploy, bootstrap, or "spin up" the Build-an-Agent workshop (a.k.a. the DevX / DevX-Lab workshop) on local hardware, or says things like "run the workshop locally", "get DevX-Lab running in my browser", "install the workshop on my GPU box", "set up the build-an-agent workshop", or "reproduce the Brev workshop deployment locally". It installs NVIDIA AI Workbench (if needed), clones and builds the project, supplies API keys, configures the GPU/host…
Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 12 other files, including scripts, reference files and assets (for example `references/access.md`, `references/persistence.md` and `references/troubleshooting.md`).
It sits in DevOps & Cloud, covering Building AI agents, Deployment and Containers. It works with NVIDIA AI Platform, Linux, Docker and Jupyter. The licence is Apache-2.0.
6 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit b5689a7. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Ships 3 files in scripts/ (Shell), which the agent can run.
Shell commands in SKILL.md call:
bashsshFrom the folder's file list and the shell code blocks in SKILL.md.
Links to these hosts (documentation or services it may open):
build.nvidia.comFrom URLs in SKILL.md, links to its own repository left out.
Names these keys or tokens, usually read from environment variables:
TAVILY_API_KEYLANGSMITH_API_KEYNVIDIA_API_KEYFrom names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Setup Workshop loads about 2.3k tokens when it runs, and up to ~5k if it reads all its reference files. Until then it costs about 217 tokens; SKILL.md has 1,069 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check noted patterns worth knowing about, such as sudo or a known installer.
are handled by setup. If sudo is not passwordless, run `sudo -v` now — theAutomated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
The full file from brevdev/workshop-build-an-agent at commit b5689a7, republished under its Apache-2.0 licence (© brevdev). 1,069 words, ~2,273 tokens.
.claude/skills/setup-workshop/SKILL.md (or your agent's skills folder). This skill also uses 9 other files; get the full folder from GitHub.Bring a local Linux NVIDIA GPU machine from bare to a browser-accessible DevX-Lab — the workshop's JupyterLab environment for all seven modules. This reproduces the Brev cloud deployment minus its cloud-only plumbing: no cloud-user detection, and the nginx single-port router + systemd auto-start are optional rather than default (local users reach the app directly via the Workbench proxy URL).
A running DevX-Lab reachable at
http://localhost:<proxyPort>/projects/<name>/applications/DevX-Lab/
(proxyPort defaults to 10000). On a desktop it can auto-open; on a headless box
it is reached via SSH port-forward.
Use on a local Linux host with an NVIDIA GPU (x86_64 or aarch64). Trigger
on requests to set up / install / deploy / spin up / run the workshop or
DevX-Lab locally. Do not use this to provision a Brev cloud instance — that
is the brev-cli / nvwb Brev path; this skill is the local analog of the Brev
startup script.
Running in an OpenShell/NemoClaw sandbox? This installer needs sudo,
Docker, and the Workbench download host — all unavailable there; it fails by
design, not by bug. Use the sandbox-native pair instead:
setup-workshop-nemoclaw (agent inside the sandbox) +
setup-workshop-nemoclaw-operator (host side: policy, secrets, port-forward).
TAVILY_API_KEY (Module 1 web search), LANGSMITH_API_KEY
(Module 3 tracing).secrets.env is gitignored, so a fresh clone never contains keys —
they must be supplied. Prompt the user for the NVIDIA key if it is not already
provided or present in an existing secrets.env.Drive these steps in order. Each script lives in scripts/. Report progress and
the final URL to the user.
Run scripts/preflight.sh. It is read-only and prints PASS/WARN/FAIL with a
verdict (exit 1 if blocking). Resolve any FAIL (disk < 40 GB, no sudo, wrong
OS/arch). GPU/docker/Workbench/network WARNs are expected on a fresh box and
are handled by setup. If sudo is not passwordless, run sudo -v now — the
background setup cannot answer a password prompt.
If no secrets.env exists for the project yet, ask the user for their NVIDIA API
key (and optionally Tavily/LangSmith). There are two ways to get it in place —
pick based on whether the project is already cloned:
secrets.env after cloning (avoids the chicken-and-egg of the file
living inside the not-yet-cloned project). See step 3.assets/secrets.env.template to <project>/secrets.env, fill in the key with
the Write tool (never echo a key in a shell command — it leaks into history
and process listings), then chmod 600 it.If a valid secrets.env already exists, no key is needed.
The first build downloads CUDA 12.8 and compiles mamba-ssm/causal-conv1d from
source — 20-45 minutes, far beyond the foreground command limit. Launch
scripts/setup.sh as a background command, passing keys via the environment:
NVIDIA_API_KEY='nvapi-…' bash .agents/skills/setup-workshop/scripts/setup.sh
# optionally also: TAVILY_API_KEY='…' LANGSMITH_API_KEY='…'
# Not in the repo? Use the symlinked path: ~/.codex/skills/setup-workshop/scripts/setup.shsetup.sh is idempotent: it starts docker, installs Workbench (if absent),
activates the local context, clones (if absent), writes secrets.env, generates
the GPU CDI spec, configures the /run/cdi/ and /var/host-run/ mounts, builds,
and starts DevX-Lab. It streams to ~/.workshop-setup.log and ends with exactly
one sentinel:
=== WORKSHOP SETUP COMPLETE ====== WORKSHOP SETUP FAILED: <reason> ===While it runs, check progress with tail -n 30 ~/.workshop-setup.log. The
harness re-invokes when the background command exits — then read the log tail to
find the sentinel.
~/.workshop-app-url for the URL and surface it with the
right access instructions for the user's situation (see step 5).references/troubleshooting.md, fix the root cause, and re-run setup.sh (it
resumes idempotently). The most common failures: missing/placeholder
secrets.env (401s), a network drop mid-build, or a missing /run/cdi source.Determine the box type and give matching instructions (full detail in
references/access.md):
start-app.sh without
--no-browser to auto-open).ssh -N -L 10000:localhost:10000 <user>@<host> — then open the URL on the
laptop.nvwb activate local --external-access then
nvwb create share-url DevX-Lab (≈48-hour public URL).Only if the user wants Brev-style behavior, offer the extras in
references/persistence.md: a systemd unit that auto-starts DevX-Lab on
reboot, and/or an nginx router that serves a clean single port (:8888) with
the Host-header rewrite. Neither is required for normal local use.
To restart the app after a reboot or to re-print the URL, run
scripts/start-app.sh (it resolves the project, starts the app idempotently, and
rewrites ~/.workshop-app-url). No rebuild needed unless package/build files
changed.
nvwb is a shell function, not a binary. Scripts must
source ~/.local/share/nvwb/nvwb-wrapper.sh (after ~/.bashrc). Never run
timeout nvwb … — timeout execs a binary and can't see the function. The
scripts here source the wrapper for exactly this reason./run/cdi/ and /var/host-run/ host mounts are declared in spec.yaml,
so their sources must exist before nvwb build/start — setup generates
the CDI spec first to avoid a cryptic "mount source not found".secrets.env / variables.env are runtime config: after editing them,
restart the container (nvwb close then nvwb open) — a rebuild is not needed.
Package/apt.txt/build-script changes do need nvwb build.--context local --project <path> explicitly instead of relying
on an "open" project, so they work non-interactively.preflight.sh — read-only environment readiness check (run first).setup.sh — idempotent end-to-end bootstrap; run in the background.start-app.sh — (re)start DevX-Lab and resolve the URL; reused by setup
and the optional systemd unit.troubleshooting.md — build failures, 401s, mounts/CDI, GPU, docker,
ports, clean-slate reset.access.md — desktop / headless port-forward / share-url / nginx access
modes and how to verify the app is up.persistence.md — optional systemd auto-start and nginx :8888 router.secrets.env.template — the keys file to fill in (NVIDIA required).nvwb-workshop.service.template — systemd unit (placeholders to substitute).nginx-workshop.conf.template — single-port router (placeholders to substitute).nvwb — full AI Workbench CLI reference (contexts, clone, build, apps,
mounts, env vars). Consult it for any nvwb command this skill invokes.nvwb-project — in-container project awareness for .project/spec.yaml;
useful once working inside DevX-Lab. See .agents/skills/VENDORED.md for
provenance and re-sync.© brevdev, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 9 other files (scripts, references, assets) in .agents/skills/setup-workshop of brevdev/workshop-build-an-agent.
Open the folder on GitHubat commit b5689a7
Setup Workshop next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Setup Workshop this skillbrevdev/workshop-build-an-agent | 144 | — | ~2.3k | Automated safety check: Notes | Apache-2.0 | |
| Investigate Production Container Restartsblotcms/blot | 2k | — | ~4.7k | Automated safety check: Pass | AGPL-3.0 | |
| Msa Search NimNVIDIA/skills | 3.5k | 1 repos | ~4.6k | Automated safety check: Notes | Apache-2.0 | |
| Tao Setup Nvidia GPU HostNVIDIA/skills | 3.5k | — | ~3.4k | Automated safety check: Notes | Apache-2.0 | |
| Frappe Ops DeploymentImpertio-Studio/Frappe_Claude_Skill_Package | 187 | — | ~2.4k | Automated safety check: Notes | MIT | |
| Headscale Deploymagnus919/agent-skills | 113 | — | ~1.3k | Automated safety check: Notes | MIT |
blotcms/blot
Work out why the blot-container-{blue,green,yellow} Docker containers from the most recent production deployment have restarted — distinguishing a normal deploy-triggered restart from a crash (V8…
NVIDIA/skills
Generate multiple sequence alignments (MSAs) for protein sequences using the ColabFold MSA-Search NIM.
NVIDIA/skills
Host setup for TAO GPU backends. An agent skill from NVIDIA/skills.
Impertio-Studio/Frappe_Claude_Skill_Package
A skill your agent uses when deploying Frappe/ERPNext to production, configuring Nginx or Supervisor, setting up Docker, enabling SSL, or hardening security.
magnus919/agent-skills
Deploy, configure, and maintain a self-hosted Headscale control server on Linux or Docker.
czlonkowski/n8n-skills
Deploys a production n8n instance to a fresh Linux server over SSH with Docker Compose and Caddy HTTPS, in single or queue mode, and covers updates, backups and hardening.
brevdev/workshop-build-an-agent
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.
brevdev/workshop-build-an-agent
Manage NVIDIA AI Workbench projects, contexts, builds, and environments via the nvwb CLI.
brevdev/workshop-build-an-agent
This skill should be used when a learner is working through Module 1 ("Build an Agent") of the Build-an-Agent workshop and wants help understanding the concepts, notebooks, or code — e.g.
brevdev/workshop-build-an-agent
This skill should be used when a learner is working through Module 1 ("Build an Agent") of the Build-an-Agent workshop and wants help understanding the concepts, notebooks, or code — e.g.
brevdev/workshop-build-an-agent
This skill should be used when a learner wants to navigate or understand the Build-an-Agent workshop as a whole — e.g.
brevdev/workshop-build-an-agent
This skill should be used when a learner wants to navigate or understand the Build-an-Agent workshop as a whole — e.g.
Works with
Categories
This skill should be used when the user wants to set up, install, deploy, bootstrap, or "spin up" the Build-an-Agent workshop (a.k.a. Setup Workshop is an agent skill from brevdev/workshop-build-an-agent.a.
Setup Workshop fits situations like: wants to set up; spin up the Build-an-Agent workshop (a.k.a.
Run `npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a claude-code`. Or copy the skill folder (.agents/skills/setup-workshop in brevdev/workshop-build-an-agent) into .claude/skills/setup-workshop in your project. Claude Code loads it when a task matches its description.
Run `npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a codex`. Or copy the skill folder (.agents/skills/setup-workshop in brevdev/workshop-build-an-agent) into .agents/skills/setup-workshop in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add brevdev/workshop-build-an-agent --skill setup-workshop -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/setup-workshop, .gemini/skills/setup-workshop, .github/skills/setup-workshop and .opencode/skills/setup-workshop in your project.
Going by SKILL.md and its folder, Setup Workshop needs a shell for the scripts in its folder, the command-line tools its instructions call (bash and ssh) and credentials named TAVILY_API_KEY, LANGSMITH_API_KEY and NVIDIA_API_KEY. Our summary lists: A Bash shell; Docker; A credential in TAVILY_API_KEY; A credential in LANGSMITH_API_KEY.
SKILL.md names 1 domain. As links in the text: build.nvidia.com. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found notes only (runs commands with sudo), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
Setup Workshop is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 2.3k tokens (SKILL.md is roughly 9.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2.7k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Setup Workshop: Investigate Production Container Restarts (blotcms/blot, 2k stars), Msa Search Nim (NVIDIA/skills, 3.5k stars), Tao Setup Nvidia GPU Host (NVIDIA/skills, 3.5k stars) and Frappe Ops Deployment (Impertio-Studio/Frappe_Claude_Skill_Package, 187 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
brevdev (a GitHub organization) maintains it in brevdev/workshop-build-an-agent, which has 144 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on October 7, 2026.
Source: brevdev/workshop-build-an-agent on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.
