A server package for self-hosting NDIF, the backend used by nnsight's remote model runs. It provides ways to start the server, run it with containers, deploy and size models, troubleshoot it, and modify its code.

It is for installing or starting NDIF with Docker, Compose, or its command-line tool, operating models, investigating problems, and developing the server itself.

It lets a team run the model-serving backend on hardware it controls instead of relying on an outside service.

ndif — Plugin for Claude Code by ndif-team/skills

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo

git clone --depth 1 https://github.com/ndif-team/skills

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/plugins/ndif-team/skills/ndif/github.svg)](https://agentmods.dev/plugins/ndif-team/skills/ndif)

Your own site

<a href="https://agentmods.dev/plugins/ndif-team/skills/ndif"><img src="https://agentmods.dev/badge/plugins/ndif-team/skills/ndif/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/plugins/ndif-team/skills/ndif"><img src="https://agentmods.dev/badge/plugins/ndif-team/skills/ndif.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session not measured What this adds to a session before it is invoked.

When invoked not measured The manifest loads nothing itself; its 4 skills cost 719 tokens a session between them.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Security

Grade A, and why

ndif scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 19d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/ndif/.claude-plugin/plugin.json · 9 lines

What it actually says

{
  "name": "ndif",
  "version": "1.0.0",
  "description": "Self-hosting NDIF, the server behind nnsight's remote=True - standing one up with docker run, compose, or the ndif CLI; deploying and sizing models; troubleshooting; and changing the server itself",
  "author": {
    "name": "NDIF Team"
  }
}

Contents

What it installs

The manifest is a name and a version. 4 skills travel with it, and installing the plugin installs all of them — 719 tokens a session between them. Each is measured on its own page, and each can be installed alone.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

19d ago First seen · 9 lines scan A 296289d7b04b

Subscribe to this mod's changes

ndif is a plugin published in the GitHub repository ndif-team/skills (14 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-15.

Cloud Containers Data and AI DevOps and cloud

gpu-server-setup

Prepare a Linux server with NVIDIA GPUs for LLM and neural-network workloads: NVIDIA driver + CUDA from the official repo, Docker Engine + NVIDIA Container Toolkit as the mandatory base, GPU passthrough into containers, staged verification gates, optional docker-compose presets for vLLM, Infinity, OpenWebUI and Ollama.

EvilFreelancer/gpu-server-setup · not measured

local-ai

From nothing to a working AI setup you own. Scans the machine for models that fit, installs and wires up the right app with consent at every download, shares it over a private URL, and rents a cloud GPU when a model is too big. Needs shell access.

naveedharri/benai-skills · not measured

agent-sandboxing

Agent sandboxing for AI coding agents: microVM and kernel-isolated execution environments. Index skill (sandbox) routes to substrate-specific paths. Currently covers the kubernetes-sigs/agent-sandbox path on kind, GKE, and kina (Kubernetes-in-Apple-Containers), with Kata, gVisor, and Apple Container microVM substrates.

vinnie357/claude-skills · not measured

ensemble-infrastructure

Infrastructure management for AWS, Kubernetes, Docker, Helm, Fly.io.

FortiumPartners/ensemble · not measured

databricks

Databricks skills for the CLI, Apps, Lakebase, Model Serving, Lakeflow Jobs, Spark Declarative Pipelines, Declarative Automation Bundles (DABs), and classic-to-serverless migration.

databricks/databricks-agent-skills · not measured

orchestra

Skills for working with Orchestra data pipelines — diagnose and fix failed runs, triage with a merge gate, author pipeline YAML, set up dbt Slim CI, configure dbt state-aware orchestration (source freshness and buildafter), run Snowflake data-quality tests, review a workspace against best practices, and merge duplicat.

orchestra-hq/orchestra-skills · not measured