doca-version
DOCA version handling: detect the installed release and validate the four-way match across pkg-config doca-common, applications/VERSION, doca_caps --version…
Skill by NVIDIA <doca@nvidia.com>
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/nvidia-doca-version
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: frontend, backend, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/doca-version
What's inside
- Skill name
doca-version- Description
- DOCA version handling: detect the installed release and validate the four-way match across pkg-config doca-common, applications/VERSION, doca_caps --version, and bfver/mlnx-release on BlueField; reason about NGC tags; look up whether a capability exists on this release; diagnose build-vs-runtime drift. Use this skill alongside doca-debug when a symptom smells of version skew: "built but nothing on the wire", "undefined reference to a symbol the docs claim exists", "DOCA_ERROR_NOT_SUPPORTED at runtime", "what does latest mean for this tag", or "is my LTS still supported". Once all four sources match, the component's build / run workflow belongs to its own library, service or tool skill. Refuse install / package choice (doca-setup), per-library API questions (matching library skill), DOCA_ERROR_* taxonomy (doca-programming-guide), and the general debug ladder (doca-debug).
- License
- Apache-2.0
- Compatibility
- No DOCA install required to read this skill (it is an overlay loaded against any DOCA artifact skill); the validation steps within DO require a live DOCA install at /opt/mellanox/doca.
- Metadata
- kind: library · author: NVIDIA <doca@nvidia.com> · doca_version: 3.5
- Instructions
- 180 lines in SKILL.md · view on GitHub
Files (7)
- BENCHMARK.md
- CAPABILITIES.md
- SKILL.md
- TASKS.md
- evals.json · evals/
- skill-card.md
- skill.oms.sig
SKILL.md (first lines)
# DOCA version ## Examples - **"What DOCA version do I actually have installed?"** — run the four-source detection chain and require a four-way match. - **"The program built but does nothing on the wire."** — treat mixed-install drift as a version problem before rewriting application code. ## Instructions Follow Loading order below. Open the tasks companion for detect / validate / diagnose, and the capabilities companion for the four-way match and NGC tag rules. Do not open every companion file. **Where to start:** This skill is the bundle's single source of truth for DOCA *version handling*. Open the tasks companion if the user wants to *do* something with the version (detect / validate / diagnose mismatch); open the capabilities companion when the question is *what does version handling cover* (the four-way match, the detection chain, NGC semantics, the per-library overlay pattern). Every other skill in the bundle that touches version routes here — they MUST NOT redefine the rules. ## Example questions this skill answers well The CLASSES of version-handling 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. - **"What DOCA version do I actually have installed?"** — worked example: *"the docs say 3.3 but I'm not sure what's on this host"*. Answered by the canonical detection chain in ## configure + ## Capabilities and modes source-of-truth table. - **"My program built but does nothing on the wire — is my install consistent?"** — worked example: *"`pkg-config --modversion` says 3.3.0; `doca_caps --version` says 3.2.0"*. Answered by the four-way match rule in ## Version compatibility + the partial-install diagnosis in ## debug. - **"Is this DOCA capability / API / sample on the version I have?"** — worked example: *"is the symmetric-RSS hash mode in Flow 2.6.0"*. Answered by the version-matrix lookup procedure in ## test (which uses the `version-matrix.json` schema defined in `doca-structured-tools-contract/SKILL.md#schemas` with fallback to per-library docs via `doca-public-knowledge-map/SKILL.md`). - **"Can I run my host package version X against BFB version Y?"** — worked example: *"host is 3.5.0 GA, BlueField BFB is 3.2.3 LTS"*. Answered by the routing to the [DOCA Compatibility Policy](https://docs.nvidia.com/doca/sdk/doca-compatibility-policy/index.html) documented in ## Version compatibility. - **"I'm inside the NGC DOCA container — what does the version match look like?"** — worked example: *"do I still need to check pkg-config / applications/VERSION / doca_caps separately?"*. Answered by the NGC container rule in ## Version compatibility + the container path in ## configure. - **"How do I write a per-library version-compatibility section for a new skill?"** — worked example: *"adding `doca-comch` to the bundle, what does its `## Version compatibility` look like?"*. Answered by the per-library overlay pattern in ## Safety policy + the worked-example template in ## modify. - **"My `apt list` shows DOCA `3.3.0109`, but `/etc/apt/sources.list.d/doca.list` is pinned at `latest` / a different release — is the next `apt install doca-*` going to silently upgrade me?"** — worked example: *"I rolled back BFB to 3.1.0105 but my sources still point at the latest channel."* Answered by the apt-source consistency precheck in ## apt-source consistency, which enumerates the three legitimate shapes of a configured DOCA apt source (network URL, local file-repo, RHEL/OEL equivalent) and the *do-not-install-until-the-source-matches* rule that protects pinned installs from accidental drift. ## When to load this skill Load this skill whenever version handling is the load-bearing concern. The decision must be made **before** the agent composes its first sentence — the activation checklist below is the same one referenced from `AGENTS.md ## Cross-cutting overlay activation triggers` (bundle `AGENTS.md#cross-cutting-overlay-activation-triggers`), mirrored here so the activation rule is at hand whenever this skill is consulted. ## Agent activation checklist — load this skill at the START of the answer when any cell below is true | Trigger class | Concrete prompt-side signals (any one fires the overlay) | | --- | --- | | Direct version question | *"what DOCA version do I have"*, *"is X consistent"*, *"is feature Y supported on version Z"*, *"can I mix host package version A with BFB version B"*, *"is my LTS still supported"*, *"what does the version string mean"* | | Container tag question | any prompt that mentions a specific NGC container tag, or asks about `latest`, or asks how to pin a tag in a Dockerfile / pod spec / Compose file. The agent MUST also cite the *"never invent a tag string from memory, never quote `latest` without confirming it"* rule from ## Safety policy. | | Build vs runtime drift | any debug session where the symptom is *"the program built fine but DOCA_ERROR_NOT_SUPPORTED at runtime"*, *"undefined reference to a symbol the docs say exists"*, *"my code does nothing on the wire"*, *"counter didn't increment"* — these are the canonical partial-install symptoms | | Upgrade / downgrade plan | the user is planning to upgrade or downgrade DOCA on a host already running other DOCA workloads, or to refresh the BFB on a BlueField pair already attached to a host | | Per-artifact cross-link | a per-artifact skill's `## Version compatibility` section cross-links here for the rule body, OR the agent is about to author a new per-artifact skill and needs the overlay template | When any cell above fires, the agent MUST: 1. Cite the **four-source detection chain** from ## Capabilities and modes explicitly in the answer — `pkg-config --modversion doca-common` → `cat /opt/mellanox/doca/applications/VERSION` → `doca_caps --version` → `bfver` plus `cat /etc/mlnx-release` (BlueField hosts). Do not paraphrase or summarize the chain; cite the commands by name. Do NOT substitute `mlxprivhost` or `bfb-info` for the BFB leg — those are common hallucinations and the bundle explicitly bans them in ## Capabilities and modes. 2. State the **four-way match rule** from ## Version compatibility verbatim if the prompt could possibly involve a mismatch (every deploy-shape question and every debug-shape question can; orientation-shape questions usually cannot). 3. Refuse to invent a version string. If the agent doesn't have the actual `pkg-config --modversion` output from the user's host, the answer must say so and route to the detection chain — not assert a version from training-data recall. ## Universal version-coherence trigger Whenever ANOTHER overlay (e.g. `doca-setup/SKILL.md`, `doca-hardware-safety/SKILL.md`, `doca-container-deployment/SKILL.md`, `doca-bare-metal-deployment/SKILL.md`) calls a `## test` / `## configure` / `## modify` step that requires *"the install is healthy"* or *"versions are consistent"*, that step MUST resolve to a citation of this skill's four-source detection chain and four-way match rule. The agent does NOT redefine the rule per-overlay — every step that needs a version verification must route here. This is the *only* place in the bundle that owns the rule body. Do **not** load this skill for general DOCA orientation, for install procedures (use `doca-setup/SKILL.md`), or for library-specific API questions (use the matching library
From NVIDIA/skills, Apache-2.0.
More from NVIDIA/skills
- 0.846score
nemo-fabric-integrate
NVIDIA Corporation and Affiliates
Use this skill when integrating NVIDIA NeMo Fabric into a consumer application, service, evaluation harness, or platform through the typed Python SDK —…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
jetson-video-pipeline
Vinit Bansal <vinitkumarb@nvidia.com>
Use when planning, executing, and independently validating Jetson Video Codec SDK or PyNvVideoCodec encode/decode, transcode, segmentation, container decode…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
nvflare-convert-pytorch
NVIDIA FLARE Team <federatedlearning@nvidia.com>
Convert existing plain or manual PyTorch training code into an NVFLARE federated job using Client API model exchange, local validation, and job export; use…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
nemo-relay-get-started
NVIDIA Corporation and Affiliates
Use this skill when first-time NeMo Relay users want help selecting and following an appropriate quick-start guide or verifying their first working integration.
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
jetson-video-recipe
Vinit Bansal <vinitkumarb@nvidia.com>
Use when turning a Jetson encoder use case into one surface-neutral recipe with native Video Codec SDK and PyNvVideoCodec projections.
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
nvidia-broadcast
NVIDIA Broadcast Team <broadcast-skills-dev@nvidia.com>
Use when controlling NVIDIA Broadcast through MCP to apply effects, process local media, select devices, or change camera resolution, or when Broadcast is…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
jetson-video-setup
Vinit Bansal <vinitkumarb@nvidia.com>
Use when installing, repairing, reusing, inspecting, or verifying readiness of the native NVIDIA Video Codec SDK or PyNvVideoCodec on Jetson, including the…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
nvflare-diagnose-job
NVIDIA FLARE Team <federatedlearning@nvidia.com>
Use when the user asks why a reported NVFLARE job failure signal occurred: the job failed, stalled, timed out, lost clients, ended with EXECUTION_EXCEPTION…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
nvflare-convert-lightning
NVIDIA FLARE Team <federatedlearning@nvidia.com>
Convert existing PyTorch Lightning training code into an NVFLARE federated job using the Lightning Client API patch, local validation, and job export; use…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
nemo-relay-install
NVIDIA Corporation and Affiliates
Use this skill when choosing or running NeMo Relay installation for an agent runtime, application language, framework integration, or plugin bundle, or…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
nvflare-fed-stats
NVIDIA FLARE Team <federatedlearning@nvidia.com>
Compute federated statistics over tabular data (count, sum, mean, stddev, var, histogram, quantile, noise-protected min/max) and image data (count…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.846score
jetson-video-capability
Vinit Bansal <vinitkumarb@nvidia.com>
Use when Jetson codec, profile, chroma, bit-depth, dimension, engine-count, or operational support must be reconciled from live APIs, authenticated NVIDIA…
SkillAFreshness: A★ 3.7KClaude Code +9
Signals
- Stars
- 3,718
- Forks
- 457
- Open issues
- 8
- Last commit (this folder)
- Oct 22, 2026 (7 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/NVIDIA/skills
- Website
- docs.nvidia.com/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from NVIDIA <doca@nvidia.com>
- 0.790score
doca-bf4-deployment
NVIDIA <doca@nvidia.com>
WARNING: guides IRREVERSIBLE BlueField-4 hardware operations (PLDM firmware burns, ISO reflashes, power cycles, BMC factory resets); every mutating step is…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.790score
doca-container-deployment
NVIDIA <doca@nvidia.com>
Running an in-bundle DOCA service as a container on a BlueField (Argus, DMS, Firefly, or UROM service): kubelet standalone watching a static-pod manifests…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.790score
doca-bf3-deployment
NVIDIA <doca@nvidia.com>
Use this skill for BlueField-3 (BF3) day-1 platform bring-up via the classic RShim/BFB path: pushing a BlueField bundle (BFB) to the DPU over RShim with…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.790score
doca-bare-metal-deployment
NVIDIA <doca@nvidia.com>
Running a DOCA-linked binary directly on the hardware, with no container or kubelet: on host x86 with a BlueField NIC over PCIe, or bare-metal on BlueField Arm.
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.790score
doca-collectx-deployment
NVIDIA <doca@nvidia.com>
Use this skill to deploy and operate a CollectX (clx) based DOCA telemetry collector on a host or BlueField — wiring providers / counters into the collector…
SkillAFreshness: A★ 3.7KClaude Code +9 - 0.790score
doca-verbs
NVIDIA <doca@nvidia.com>
Raw verbs inside DOCA Core (doca-verbs): direct control of QP, CQ, PD, SRQ and address-handle objects through doca_verbs_context and doca_verbs_qp_create /…
SkillAFreshness: A★ 3.7KClaude Code +9
Cite this page
doca-version skill. Agent Mart. https://agentmart.dev/skills/nvidia-doca-version (last updated 2026-10-24)