Skip to main content
affpapa.org
RU EN
skill CX Codex

nemotron-policy-generator — Claude Code skill for safety policies

Turn moderation requirements or an existing policy into structured artifacts for NVIDIA Nemotron safety models. Useful when a project needs its rules prepared for model inference.

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

What nemotron-policy-generator does

The skill turns user-provided moderation needs into a Markdown policy, a JSON taxonomy, and a system prompt for classification. It maps plain-language rules to V2 categories and can add project-specific categories or topic-following constraints. A DOCX version is produced when requested, including for sign-off or review.

Its outputs target nvidia/Nemotron-Content-Safety-Reasoning-4B and multimodal nvidia/Nemotron-3-Content-Safety. These models use different labels and output modes, so the taxonomy carries metadata for both. Nemotron-3 handles text and images; Reasoning-4B is text-only and English-language. The default is a policy artifact for both models, so name one model if the project targets only that one.

What it helps with

  • turn brief moderation requirements into a structured policy
  • map project rules to V2 safety categories
  • add custom categories to a safety taxonomy
  • prepare a classification prompt for Nemotron models
  • draft a labeling rubric for training data

Example prompts for the agent

  • Draft a chat policy that blocks weapon instructions while allowing medical discussion.
  • Convert this moderation policy into V2 categories and add a rule for our subject area.
  • Create a policy for nvidia/Nemotron-3-Content-Safety that classifies text and images.

How to install nemotron-policy-generator

1Skills CLI (Vercel)

npx skills add NVIDIA/skills --skill nemotron-policy-generator -a claude-code

2Manually: put the files in place

~/.claude/skills/nemotron-policy-generator/SKILL.md

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

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

Questions about nemotron-policy-generator

How do I install nemotron-policy-generator in Claude Code?

Run npx skills add NVIDIA/skills --skill nemotron-policy-generator -a claude-code. The skill file will be at ~/.claude/skills/nemotron-policy-generator/SKILL.md.

What does nemotron-policy-generator do?

It creates a Markdown policy, JSON taxonomy, and classification system prompt from moderation requirements or an existing policy. It can also add custom categories and topic constraints.

Which models does the skill support?

It prepares artifacts for text-only nvidia/Nemotron-Content-Safety-Reasoning-4B and multimodal nvidia/Nemotron-3-Content-Safety. By default, the policy targets both.

Similar skills

All skills for AI and ML →
skill-creator official anthropics Creates and refines skills, then tests them with prompts, evaluations, and performance checks. 414,180installs writing-for-agents Matt Pocock Guidance for writing skills and agent-facing documents. Use it when editing AGENTS.md, CLAUDE.md, or reference files reached from them. 466,719installs media-use heygen-com Finds or generates media for HyperFrames videos, including voiceovers, music, cutouts, and color grades. Stores files locally in assets. CX Codex +1 1,100,000installs mcp-builder official anthropics This skill walks through building an MCP server in TypeScript or Python, from reading the protocol spec to writing tests and evaluation questions. 130,949installs subagent-driven-development Jesse Vincent This skill runs an implementation plan through fresh subagents for independent tasks and reviews each result. Use it for multi-step work in one session. CX Codex 237,746installs orchestration stablyai Coordinates supervised Orca workers with task DAGs, decision gates, and threaded messages. Fits workflows that split work across multiple agents. CX Codex +1 349,800installs

From the same set195

digital-health-clinical-asr-setup Sets up the environment for NVIDIA's Clinical ASR Flywheel, verifies the NVIDIA_API_KEY, installs Python dependencies, and runs an end-to-end smoke test. CX Codex +1 1,900installs dicom-series-to-volume Turns one CT DICOM series folder into an HU NIfTI volume plus a JSON summary with the affine. Handy before feeding scans to a segmentation model. CX Codex +1 1,900installs dicom-series-preflight Reads DICOM headers in a single series folder and returns JSON with a pass, warn or fail verdict before conversion or inference. CX Codex +1 1,900installs dicom-metadata-extract Runs a pydicom script on a single DICOM file to extract metadata and detect standard PHI tags. Useful for quick agent inspections of medical imaging files. CX Codex +1 1,900installs holoscan-install-source Builds and verifies NVIDIA Holoscan SDK 5.x from source using the repository container wrapper. Helps with Early Access branches and custom SDK components. CX Codex +1 1,800installs holoscan-install-debian Installs native C++ Debian packages of Holoscan SDK 5.x on Ubuntu 24.04 and checks that the bundled examples build. CX Codex +1 1,800installs holoscan-install-wheel Installs Holoscan SDK 5.x CUDA 13 Python wheels into an isolated venv and checks that the import works. CX Codex +1 1,800installs holoscan-install-conda Checks Conda package availability for Holoscan SDK 5.x releases, flags missing 5.0 builds, and routes to supported installation alternatives. CX Codex +1 1,800installs

Project sponsors

Companies that keep this analytics open
Gold sponsor
Nutra.Club

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