NVIDIA/skills is a catalogue of portable instruction sets that teach coding agents how to use NVIDIA software for robotics, simulation, CUDA, retrieval-augmented generation, and related workflows. Developers install these skills in agents such as Claude Code or Codex, while the catalogue mirrors skills maintained in separate NVIDIA product repositories.
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 skills add NVIDIA/skills --skill doca-pccgit clone --depth 1 https://github.com/NVIDIA/skillsWrote 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/skills/doca-pcc)<a href="https://agentmods.dev/skills/nvidia/skills/doca-pcc"><img src="https://agentmods.dev/badge/skills/nvidia/skills/doca-pcc.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 5.1 | $0.00240 | $0.04899 |
| Opus 5 | $0.00120 | $0.02449 |
| Sonnet 5 | $0.00048 | $0.00980 |
| Haiku 4.5 | $0.00024 | $0.00490 |
Grade A, and why
doca-pcc scanned grade A with 0 findings 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 3d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 372 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DOCA PCC
Where to start: This skill assumes DOCA is already installed,
the user's BlueField has a DPA processor that the host can see
through DOCA, the BlueField firmware has the custom-PCC slot
enabled, and the user is doing hands-on custom PCC work from
the host side — i.e. using doca-pcc to load a DPA-side
congestion control algorithm onto the BlueField, attach it to a
port handling RDMA / RoCE traffic, and parameterize it from the
host. Open TASKS.md if the user wants to do
something (configure / build / modify / run / test / debug);
open CAPABILITIES.md when the question is
what can the host-side PCC API express on this version + this
BlueField generation + this firmware. If the user has not
installed DOCA yet, route to
doca-setup first; if the user is
asking how to write the DPA-side congestion-control algorithm
itself (the code that runs on the DPA processor, compiled by
dpacc), that is a different scope — route via
doca-public-knowledge-map
to the public DOCA PCC programming guide and to
doca-dpa for the host-side DPA
lifecycle this skill builds on. If the user only wants to
inspect PCC counters at runtime without writing a custom
algorithm, that is the pcc_counters CLI tool — route via
doca-public-knowledge-map ## DOCA tools;
this skill is for custom algorithms only.
Example questions this skill answers well
The CLASSES of PCC questions this skill is built to answer, each with one worked example. The agent should treat the class as the load-bearing piece — the worked example is a single instance.
- "How do I deploy my own custom congestion control algorithm
onto a BlueField port carrying RDMA / RoCE traffic?" —
worked example: "load a small DPA-side PCC algorithm and
attach it to the BlueField port that handles my RoCE traffic".
Answered by the two-side-program model + the host-side
load-and-attach workflow in
CAPABILITIES.md ## Capabilities and modes- the bring-up steps in
TASKS.md ## configure.
- the bring-up steps in
- "Does this BlueField + firmware actually allow a custom PCC
algorithm, and which PCC features does my DOCA install
expose?" — worked example: "my host has a BlueField and the
default factory PCC works; can I drop in a custom algorithm
instead?". Answered by the triple-axis precondition rule
(BlueField generation must carry a DPA, firmware must have the
custom-PCC slot enabled,
doca_pcc_cap_*against the activedoca_devinfomust agree) inCAPABILITIES.md ## Capabilities and modes- the env-precondition checklist in
TASKS.md ## configurestep 1.
- the env-precondition checklist in
- "Why does my custom PCC fail with
DOCA_ERROR_NOT_PERMITTEDeven though I havedoca_devaccess?" — worked example: "the BlueField firmware in this host has the custom-PCC slot disabled". Answered by the permission matrix inCAPABILITIES.md ## Safety policy- the firmware-side env fix in
TASKS.md ## configurestep 1.
- the firmware-side env fix in
- "Is this
doca-pcclibrary the right tool for what I want, or do I want the default firmware PCC or thepcc_countersCLI?" — worked example: "I just want to read PCC counters without touching the algorithm". Answered by the path-selection rule inCAPABILITIES.md ## Capabilities and modes- the deferred-topic boundaries in
CAPABILITIES.md ## Deferred topic boundarieswhich route todoca-public-knowledge-map ## DOCA toolsfor the counter tool.
- the deferred-topic boundaries in
- "Is the host-side PCC API I'm reading about on my installed
DOCA?" — worked example: "is the host-side load helper I
see in the docs available against the DOCA + DPACC versions on
this host?". Answered by the version-compatibility overlay in
CAPABILITIES.md ## Version compatibilitywhich cross-links the canonical detection chain indoca-versionand adds the PCC-specific DOCA must match DPACC overlay inherited fromdoca-dpa. - "What does this
DOCA_ERROR_*from adoca_pcc_*call mean and which layer caused it?" — worked example: "DOCA_ERROR_DRIVERon the host-side algorithm-load call — is it DOCA, the firmware-side custom-PCC slot, or the DPACC- produced image?". Answered by the PCC overlay on the cross-library taxonomy inCAPABILITIES.md ## Error taxonomy- the layered ladder in
TASKS.md ## debugthat escalates todoca-debug.
- the layered ladder in
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.
- 3d ago First seen · 372 lines · 240 tokens per session scan A e74ce6abe8e5
doca-pcc is a skill published in the GitHub repository NVIDIA/skills (3,223 stars, last pushed today), licensed Apache-2.0. It adds 240 tokens to every session and 4,899 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
terraform-module-library
Build reusable Terraform modules for AWS, Azure, GCP, and OCI infrastructure following infrastructure-as-code best practices. Use when creating infrastructure modules, standardizing cloud provisioning, or implementing reusable IaC components.
hyperpod-version-checker
Check and compare software component versions on SageMaker HyperPod cluster nodes - NVIDIA drivers, CUDA toolkit, cuDNN, NCCL, EFA, AWS OFI NCCL, GDRCopy, MPI, Neuron SDK (Trainium/Inferentia), Python, and PyTorch. Use when checking component versions, verifying CUDA/driver compatibility, detecting version mismatches…
technical-troubleshooting
Provide setup, troubleshooting, and maintenance guidance. Use when the user reports a device that won't power on, connectivity issues, setup questions, overheating, or maintenance concerns.
gpu-optimizer
GPU optimization for consumer NVIDIA GPUs (8-24GB VRAM) covering mixed precision, gradient checkpointing, XGBoost GPU, CuPy/cuDF migration, and torch.compile. Triggers on: "optimize GPU training", "speed up CUDA", "reduce OOM", "migrate NumPy to CuPy", "manage GPU memory", "benchmark PyTorch".