Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/nvidia/nvcf/nvcf-self-managed-installationnpx skills add NVIDIA/nvcf --skill nvcf-self-managed-installationgit clone --depth 1 https://github.com/NVIDIA/nvcfWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nvidia/nvcf/nvcf-self-managed-installation)<a href="https://agentmods.dev/skills/nvidia/nvcf/nvcf-self-managed-installation"><img src="https://agentmods.dev/badge/skills/nvidia/nvcf/nvcf-self-managed-installation.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00110 | $0.06245 |
| Opus 5 | $0.00055 | $0.03122 |
| Sonnet 5 | $0.00022 | $0.01249 |
| Haiku 4.5 | $0.00011 | $0.00624 |
Grade A, and why
nvcf-self-managed-installation scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -H "Host: vanity.<domain>" "http://<gateway-address>/health" How it starts
The opening of the file, as written. The whole thing — 500 lines — stays where its author put it; the contents beside it link to each section on GitHub.
NVCF Self-Managed Stack Operations
Operational guide for matching NVCF control-plane and compute-plane Helmfile bundles.
Instructions
Use this skill for install, upgrade, or teardown work in matching nvcf-self-managed-stack and nvcf-compute-plane-stack bundles; keep Helmfile values and CLI profile handoffs separate. For functionType: "LLM", read LLM Function Enablement.
Prerequisites
Ask for the extracted nvcf-self-managed-stack path. For split installation,
also require the matching-version nvcf-compute-plane-stack path. Verify both:
ls <control-plane-stack>/helmfile.d/ <control-plane-stack>/environments/ <control-plane-stack>/secrets/ <control-plane-stack>/global.yaml.gotmpl
ls <compute-plane-stack>/helmfile.d/ <compute-plane-stack>/environments/ <compute-plane-stack>/global.yaml.gotmpl
If either directory is missing, download matching bundle versions from NGC:
ngc registry resource download-version <org>/nvcf-self-managed-stack:<version>
ngc registry resource download-version <org>/nvcf-compute-plane-stack:<version>
Control-plane commands below assume the control-plane bundle root. Split-flow commands and root substitution are in Split Compute-Plane Installation.
Before You Start
Verify tooling and context before any operation:
helmfile --version # Must be 1.1.x (1.2.0 removed sequential mode)
helm version # Must be >= 3.12
helm plugin list # Must include helm-diff >= 3.11
kubectl version # Client must be within 1 minor version of cluster
Control-plane commands run from the extracted nvcf-self-managed-stack/ root:
cd path/to/nvcf-self-managed-stack
ls helmfile.d/ environments/ secrets/ global.yaml.gotmpl
Identify the environment from environments/<name>.yaml and
secrets/<name>-secrets.yaml; for EKS, use the canonical
CSP End-to-End Example.
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 500 lines · 110 tokens per session scan A c33fd8305ec3
nvcf-self-managed-installation is a skill published in the GitHub repository NVIDIA/nvcf (202 stars, last pushed 4d ago), licensed Apache-2.0. It adds 110 tokens to every session and 6,245 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal
Use when running Python or GPU workloads serverlessly on Modal — modal.App, inline container Images, gpu= on @app.function, Volumes for weight caching, Cron schedules, ASGI endpoints, modal run vs serve vs deploy. NOT managed prediction APIs with no container of your own (that is replicate); NOT SSH-able GPU boxes…