Agent skill · software engineering · nvidia
doca-bf4-deployment
WARNING: guides IRREVERSIBLE BlueField-4 hardware operations (PLDM firmware burns, ISO reflashes, power cycles, BMC factory resets); every mutating step is governed by doca-hardware-safety (window, rollback). BlueField-4 only: BlueField-3 bring-up is doca-bf3-deployment. Use this skill for BlueField-4 (BF4) day-1 platform bring-up from the BMC: installing the BlueField/DOCA bundle ISO onto the DPU (Grace, the Arm complex) over UEFI HTTP Boot, PXE, or Redfish Virtual Media; the PLDM platform-firmware update via the Redfish UpdateService and pldmtool; and a Grace Ubuntu image with optional cloud-init. Trigger on BlueField-4/BF4 bring-up phrasings even without "BF4": {bring up my new BlueField-4}, {the BlueField ISO will not boot over HTTP from the BMC}, {attach BF4 virtual media via Redfish}, {BF4 firmware Task stuck at Running}, {prepare a BlueField-4 host: platform support, version alignment, safest first smoke test}. Route application launch and library APIs to other skills.
Why this skill is useful
Adds detailed executable workflows for BlueField-4 hardware operations, including firmware updates and OS installations that are not covered in public documentation.
What it needs
Requires pldmtool installed locally. About 7k tokens when loaded. Last updated 2026-08-07. 2,816 stars on the source repository.
What this skill does
DOCA BlueField-4 (BF4) deployment Examples "I have a BlueField-4 and a BMC — how do I bring Grace up?" — BMC-driven day-1, not the BF3 RShim/BFB path. Do not invent credentials or skip media detach. "PLDM update finished but the NIC firmware did not move." — treat it as a firmware-Task / activation failure, not a second burn by default. BlueField-3 or any other generation is not this skill. Name doca-bf3-deployment and stop; do not read further or run commands. Preparing a BlueField-4 host (answer from here) 1. Platform: confirm it is BlueField-4 and that the BMC is reachable out of band; compare the Redfish firmware inventory with the release notes' targets. 2. Version alignment: run the doca-version four-source chain on the host. Fixing any mismatch changes hardware state and follows doca-hardware-safety. 3. Safest first smoke test: read-only docacaps --list-devs, then one shipped sample per doca-bare-metal-deployment. No firmware, mode or reset change before that is green. 4. Say it in the answer: every mutating step (PLDM firmware burn, ISO reflash, power cycle, BMC reset) is routed through doca-hardware-safety: pre-flight inventory, out-of-band access, maintenance window, rollback, then apply and verify. Instructions Follow Loading order below. Open the tasks companion for BMC bring-up and recover. Open the capabilities companion for the three install methods and the BF4 safety overlay. Do not open every companion file. ⚠️ WARNING — irreversible hardware operations. This skill guides operators through potentially destructive, irreversible BlueField-4 hardware operations: PLDM firmware burns, ISO reflashes, power cycles, and BMC factory resets. These can brick firmware, corrupt boot media, or cause production outages. Do not proceed without a maintenance window and a tested rollback plan. Every mutating step is governed by doca-hardware-safety/SKILL.md, which MUST be loaded alongside this skill before any destructive action. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills nvidia/doca-bf4-deployment