NVEIL is a local-first data processing and visualization toolkit that uses a constraint-solver to execute complex data science tasks via natural language.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install nveil
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install nveil using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is NVEIL?
NVEIL is a revolutionary Python toolkit designed for Openclaw Skills that eliminates the need for manual data science coding. Instead of writing fragile pandas pipelines, Plotly configurations, or complex VTK scene setups, you simply describe your data goals in plain language. The system utilizes a remote constraint-solving engine to plan the execution, which then runs entirely on your local machine to ensure your raw data never leaves your environment.
By serving as a deterministic and auditable agent for data science, NVEIL provides a decisive advantage for LLM-driven workflows. It collapses the cycle of guessing column names and debugging runtime errors into a single, reliable call. Whether you are dealing with tabular data, 3D medical imaging, or geospatial maps, NVEIL handles the backend complexity automatically, choosing the best library for the task without manual intervention.
NVEIL Use Cases
- Automating complex data joins, aggregations, and pivots across multiple datasets using natural language.
- Creating professional 2D and 3D visualizations, including heatmaps, Sankey diagrams, and voxel volumes.
- Performing time-series analysis with resampling and rolling statistics for financial or sensor data.
- Visualizing medical imaging files like DICOM and MHD locally and securely.
- Geocoding city names and coordinates for interactive choropleth and point-cloud maps.
How NVEIL Works
- Inspection: The toolkit first describes the dataset to understand shapes, dtypes, and schemas locally.
- Planning: A natural language prompt, along with the schema and summary statistics, is sent to the NVEIL planning service.
- Constraint Solving: The server generates a deterministic execution plan (spec) based on the metadata provided.
- Local Execution: The NVEIL Python package executes the plan on the user's machine, processing the raw data and rendering outputs.
- Output Generation: The process concludes by generating files such as HTML, PNG, or reusable .nveil specs for offline use.
NVEIL Setup
To get started with this skill for Openclaw Skills, install the Python package and configure your API key:
pip install nveil
export NVEIL_API_KEY='your_api_key_here'
For 3D scientific rendering (VTK-backed), use the extra installation:
pip install 'nveil[extra]'
NVEIL Data Schema & Taxonomy
NVEIL organizes data processing and metadata through a structured local workflow. It supports various input formats and generates portable specification files.
| Feature | Supported Formats / Details |
|---|---|
| Input Tables | CSV, Parquet, JSON, Excel, pandas DataFrames |
| Input Imaging | DICOM, MHD (Medical), PNG, JPG, TIFF (Images) |
| Generated Specs | .nveil files (portable, replayable, and deterministic) |
| Output Types | Interactive HTML, PNG (via headless Chromium), and CLI summaries |
NVEIL Advanced Features
- Local Render Engine: Use
nveil renderto replay existing specs on fresh data without any API calls or network costs. - MCP Integration: Includes a Model Context Protocol (MCP) server for seamless integration with Claude Desktop, Cursor, and other AI IDEs.
- Multi-Dataset Joins: Ability to pass a dictionary of multiple files or DataFrames for complex relational operations.
- Customizable Backends: Automatically selects and configures high-performance backends like VTK, DeckGL, and ECharts-GL.
- Privacy-First Architecture: Only metadata and summary statistics are transmitted, ensuring raw data privacy for finance and healthcare.
SKILL.md
Loading
Related Openclaw Skills
Alibaba Cloud DataWorks Data Quality (Read-Only)
A specialized read-only interface for querying, auditing, and diagnosing Alibaba Cloud DataWorks Data Quality monitors and execution logs.
Statamic AI Gateway
A secure, authenticated tool execution gateway for managing content across multiple Statamic sites using AI agents.
Umeng+ App Data Analysis
A comprehensive integration for Openclaw Skills that enables developers to fetch and analyze mobile application performance data directly from the Umeng+ Open API.
FMTWiki
A specialized clinical knowledge base skill for managing Fecal Microbiota Transplantation data through multi-agent verification and evidence-based tracking.
Speechmatics
Transcribe audio files like podcasts, meetings, and voice notes into text, JSON, or subtitles via the Speechmatics API.
Video Generator Free Models
A high-performance AI skill that transforms text prompts into high-definition 1080p videos using free cloud-based models.
Featured*
