What does this skill do, and when should you use it?
This skill bootstraps first-run customization for a custom carrier in the Jetson Linux BSP. It forks and renames carrier-related configuration, creates a device-tree overlay, and connects nvpmodel, nvfancontrol, and nvpower.sh behavior to the custom carrier. The kernel base DTB is not forked; carrier changes are delivered through OVERLAY_DTB_FILE. It requires jetson-init-source first and an active profile containing custom_carrier.
Resolves the active target, reference module, carrier identifiers, and BSP paths; checks the Linux_for_Tegra Git tracker and reference BSP image; discovers and derives the board flash configuration, MB1/MB2 BCT files, nvpmodel, and nvfancontrol files; performs targeted flash-conf rewrites; creates the Jetson device-tree plugin overlay and Makefile registration; appends OVERLAY_DTB_FILE; patches nvpower.sh; re-verifies edits after patches; and previews staged files and proposed commit messages before each commit. BPMP DTB derivation is optional.
- A Jetson BSP engineer is bringing up a new custom carrier derived from a reference devkit.
- An embedded team needs to preserve the reference kernel base DTB while layering carrier-specific device-tree changes.
- A carrier bring-up project needs coordinated boot configuration, BCT, power-model, fan-control, and nvpower.sh changes.
- A team is re-deriving a carrier customization against a newer reference BSP and wants the carrier-specific fileset regenerated consistently.
How do you install this skill?
- The skill modifies multiple BSP files, creates commits, and instructs pushing to origin; verify the repository, staged content, commit messages, and push authority, and maintain an independent backup or rollback path.
- The skill depends on several context and reference files whose contents were not supplied; this review did not execute commands and cannot confirm paths, substitutions, or overlay builds on a specific BSP.
- The skill-card license statement is not fully consistent with the SKILL.md frontmatter, and the benchmark reports an author-format warning; governance metadata should be reconciled before publication.
- No Chinese documentation or mainland-China network reachability evidence is provided.
- Shell / CLI
- Network access
- Local filesystem
GitNVIDIA Jetson Linux BSP source treeresolved active profileinitialized Linux_for_Tegra overlay trackerpopulated reference BSP image
Install it with the skills CLI: npx skills add nvidia/skills --skill jetson-derive-carrier --yes. The README does not document a separate manual installation directory for this skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/NVIDIA/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/jetson-derive-carrier" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- derive custom carrier
- bootstrap custom carrier
Run jetson-init-source first, then ensure the active profile provides custom_carrier, reference_devkit, source.root_path, and bsp_image.root_path. Prompt the agent with “derive custom carrier” or “bootstrap custom carrier”. The skill refuses to run when custom_carrier is absent or Linux_for_Tegra/.git is missing, and it requires operator approval at each commit-message preview gate.
What are this skill's strengths and limitations?
- Covers the carrier derivation fileset across flash configuration, BCT, device-tree overlay, power model, fan control, and nvpower.sh.
- Keeps the reference kernel base DTB intact and wires carrier deltas through an overlay.
- Defines a single discovery pass, row-by-row fork rules, edit verification, and commit preview gates.
- Accepts custom carrier identifiers that do not use NVIDIA’s pNNNN format.
- It does not handle module-level customization or kernel-DTB changes.
- It depends on a resolved active profile, a populated reference BSP image, and an initialized Linux_for_Tegra Git tracker.
- BPMP DTB derivation is off by default and must be selected when the carrier’s power or clock topology diverges.
- The supplied source does not provide independent test results or a platform support matrix for this individual skill.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Jetson Custom Carrier Derivation this page ✓ NVIDIA · Official | 53 · Use with care | ★ 3.8k | today | Apache-2.0 |
| Jetson Fan Curve Customizer ✓ NVIDIA · Official | 54 · Use with care | ★ 3.8k | today | Apache-2.0 |
| Jetson Thor MGBE/QSFP Configuration ✓ NVIDIA · Official | 49 · Use with care | ★ 3.8k | today | Apache-2.0 |
| Jetson USB Port Customization ✓ NVIDIA · Official | 55 · Use with care | ★ 3.8k | today | Apache-2.0 |
| Jetson BSP Source Builder ✓ NVIDIA · Official | 58 · Recommended | ★ 3.8k | today | Apache-2.0 |
How did FollowSkills review this skill?
The skill specifies prerequisites, a single discovery pass, operator confirmation before commits, preservation of the staged index on cancellation, and avoidance of staging reference files; no credential theft, covert exfiltration, or malicious dependency is shown. Deductions apply because it instructs pushing to origin without a separate explicit push confirmation, provides limited rollback guidance, and relies on context files not included in the evidence.
The workflow, naming rules, warnings, post-edit verification, and warn-and-skip behavior are detailed. Deductions apply because there is no implementation script or key-path test suite, static review cannot reproduce behavior, and several paths depend on external context and BSP-specific layouts, leaving some diagnostic burden with the operator.
Triggers, audience, prerequisites, non-fit scope, input fields, and output files are reasonably clear, and non-pNNNN carrier IDs are supported. Deductions apply because there is no Chinese-language or mainland-China network-fit evidence, and semantic trigger boundaries are supported mainly by a small example set.
The skill has good structure, versioning, dependency notes, limitations, troubleshooting, and examples. Deductions apply because the skill card's license statement is not fully consistent with the frontmatter, the benchmark reports an author-format finding, and no clear changelog, stable parameter contract, or concrete update path is provided.
The intended goal and outputs are clear, covering carrier-file derivation, DT overlay creation, Makefile registration, and power-configuration integration, which could reduce manual omissions. Deductions apply because actual outputs were not verified statically, the benchmark contains only two positive tasks, and its reported effectiveness is limited; BSP variation may still require substantial manual review.
The fixed revision, evaluation report, evaluation tasks, and ground truth provide some auditability. Deductions apply because there is no independently reproducible committed test suite or CI covering the skill's key paths; the small evaluation sample cannot substitute for execution evidence.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →