Skip to main content
skill CX Codex

skill-card-generator — Claude Code skill: governance cards

Turns a skill folder that already exists into a draft governance card: it reads the files, writes a context, renders markdown and checks that review markers are gone.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add NVIDIA/skills --skill skill-card-generator -a claude-code
Open on GitHub

What skill-card-generator does

Point it at a skill directory and it produces a draft governance card: what the skill does, which permissions it asks for, what it needs from the machine. The path comes in as an argument. No argument means the current folder.

Three Python scripts carry the work. One scans the target and hands back a report on its files and repo metadata. From that report the agent writes a context JSON, following the per-field hints in references/style-guide.md. A Jinja template at references/skill-card.md.j2 turns the JSON into markdown, and the renderer validates the shape of the context on the way through.

Reads stay inside the target folder plus this skill's own references/ and scripts/. Writes land in the target or /tmp/, and shell calls are limited to the three named scripts. jinja2 must be installed before rendering, and Python 3 is needed from the start. Output is a draft — the validator only fails when VERIFY or SELECT markers survive in the file.

What it helps with

  • draft a governance card for an existing skill folder
  • refresh a card after the skill itself changed
  • pull file and repo metadata out of the target folder
  • catch leftover VERIFY and SELECT markers before submitting
  • get material ready for legal and safety review

Example prompts for the agent

  • Generate a governance card for the skill in ./skills/my-skill
  • Refresh the card for /path/to/target-skill, SKILL.md has changed
  • Build a draft card for the current folder and run the marker check

How to install skill-card-generator

1Skills CLI (Vercel)

npx skills add NVIDIA/skills --skill skill-card-generator -a claude-code

2Manually: put the files in place

~/.claude/skills/skill-card-generator/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/skill-card-generator/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about skill-card-generator

How do I install skill-card-generator in Claude Code?

Run npx skills add NVIDIA/skills --skill skill-card-generator -a claude-code. The skill lands at ~/.claude/skills/skill-card-generator/SKILL.md.

What does skill-card-generator do?

It builds a draft NVIDIA governance card for a skill directory that already exists, using signals it discovers inside that directory.

What has to be installed first?

Python 3, the jinja2 package for the renderer, and a target folder containing SKILL.md or skill.md.

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set194

deepstream-import-vision-model This skill takes an object-detection model from HuggingFace or NGC through a full NVIDIA DeepStream pipeline: ONNX, TRT engine, bbox parser, benchmark, PDF report. CX Codex +1 2,300installs omniverse-usd-performance-tuning A skill for diagnosing and optimizing USD scenes: slow loading, low FPS, high memory. It owns the user-facing workflow across phases. CX Codex +1 2,300installs cuopt-server-api-python Deploys the NVIDIA cuOpt REST server and guides client requests via Python or curl. Useful for solving routing, LP, and MILP optimization problems on GPUs. CX Codex +1 2,200installs rag-perf Runs a profiling pass and an aiperf load test against a deployed NVIDIA RAG Blueprint server, all driven by one YAML config. CX Codex +1 2,200installs cuopt-numerical-optimization-formulation Turns a written problem into a clean LP, MILP or QP formulation: variables, constraints, objective. Concepts only, no API code. CX Codex +1 2,100installs dali-dynamic-mode A skill that steers the agent through NVIDIA DALI's imperative mode: direct ndd calls instead of pipeline graphs, plus help migrating existing pipelines. CX Codex +1 2,100installs nemotron-customize Plans DAG pipelines and generates YAML configs for Nemotron customization steps, from data curation and SFT to RL alignment and evaluation. CX Codex +1 2,100installs physical-ai-neural-reconstruction A thin router for NVIDIA NuRec work: it points you to the right skill for converting a recording, training a reconstruction or rendering a USDZ. CX Codex +1 2,100installs

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor