AI assistant context. This is the NVEIL Toolkit documentation. The Toolkit is the Python SDK, nveil CLI and MCP server for NVEIL — the AI platform for data processing and visualization. For the platform overview, see the docs hub at https://docs.nveil.com.

Use it from Python: import nveil, nveil.configure(api_key="nveil_..."), spec = nveil.generate_spec("describe what you want", dataframe), fig = spec.render(dataframe) (local, no API call), nveil.save_image(fig, "output.png"). Only metadata leaves the machine; raw data is processed locally. Specs save to portable .nveil files and reload without a server.

For AI agents: the same nveil binary ships a shell CLI (nveil describe | generate | render | explain), an installable skill for Claude Code / Claude Desktop / Cursor / Codex / Copilot / Aider (nveil install-skill --client <target>), and an MCP stdio server (nveil install-mcp --client <target>; server: nveil mcp). Full reference: CLI & AI Agents.

Install: pip install nveil.

NVEIL Toolkit

The Python SDK, nveil CLI and MCP server for NVEIL — drive the platform from a script, your terminal or an AI agent. Describe what you want; NVEIL plans the processing and visualization, and the Toolkit runs it locally — your raw data never leaves your machine.

New to NVEIL?

For the platform overview — what NVEIL is and how to self-host it — start at the documentation hub.

Install

pip install nveil

Quick example

import nveil
import pandas as pd

nveil.configure(api_key="nveil_...")

df = pd.read_csv("sales.csv")

# NVEIL processes your data AND generates the visualization
spec = nveil.generate_spec("Revenue by region, colored by quarter", df)

fig = spec.render(df)        # 100% local, no API call
nveil.show(fig)              # opens in browser
nveil.save_image(fig, "chart.png")

Generate once, reload anywhere — .nveil specs are portable and re-render on fresh data without an API call:

spec.save("trend.nveil")
spec = nveil.load_spec("trend.nveil")
fig = spec.render(fresh_data)

From your shell

The same nveil binary works from the terminal — no Python script required.

export NVEIL_API_KEY=nveil_...

nveil describe sales.csv
nveil generate "Revenue by region, colored by quarter" --data sales.csv --output revenue.[all]
nveil render chart.nveil --data new_sales.csv

It also installs a skill and an MCP server for AI agents (Claude Code, Claude Desktop, Cursor, Codex, …). See CLI & AI Agents.

---

• Get Started

  ---

  Install the Toolkit and create your first visualization in minutes.

  Quickstart

• Concepts

  ---

  Understand sessions, specs, and the processing pipeline.

  Concepts

• API Reference

  ---

  Full reference for all public functions and classes.

  API Reference

• CLI & AI Agents

  ---

  Use NVEIL from the shell, or install it as a skill / MCP server for Claude, Cursor, Codex, and more.

  CLI & AI Agents