csvglow is an open-source project whose goal is to let users generate a beautiful, interactive HTML dashboard from a CSV or Excel file with a single command and zero configuration. The generated dashboard opens in the browser and includes automatically detected charts, intelligent multi-column insights, correlation analysis, and a sortable data table. The dashboard uses a dark gradient theme, and every chart provides a copy button for easy pasting into slides.
Key capabilities
- Smart discovery: narrative analysis across multiple columns, cross-referencing metrics to reveal contradictions, efficiency gaps, and best/worst performers.
- Histograms: generate a histogram for each numeric column, with mean, median, standard deviation, quartiles, and outlier counts displayed.
- Bar charts: generate bar charts for categorical columns.
- Cross analysis: automatically generate category x numeric cross-tabulations with an overall mean line.
- Time series charts: generate line charts with area fill for date columns.
- Correlation heatmap: show correlations between numeric columns.
- Scatter plots: generate scatter plots for highly correlated pairs (|r| > 0.7).
- Data table: sortable, filterable, showing the first 1000 rows.
- Copy buttons: every chart has a copy button for easy pasting into slides.
The output is a single, self-contained HTML file. No server, no CDN, works offline.
Usage
- Command line: after installation, one command generates the dashboard. You can also run it directly with npx without installing.
- MCP tool: csvglow can be used as an MCP tool in any MCP-compatible client. Once configured, you can ask the AI assistant to generate a dashboard from a file path. Add csvglow to the client's MCP configuration file; using npx requires no additional installation. If csvglow has been installed via pip, you can use "command": "csvglow" with "args": ["--mcp"].
- OpenClaw skill: csvglow is also available as an OpenClaw skill. Any OpenClaw-compatible client can automatically discover and use it without manual configuration.
Supported formats
- .csv / .tsv (automatic delimiter detection)
- .xlsx (first worksheet only—multi-worksheet support coming soon)
Other information
- Initial version includes: automatic detection of column types (numeric, categorical, datetime, identifier); smart discovery: cross-column contradiction detection, efficiency analysis, best/worst performer identification; histograms (with statistics sidebar), bar charts, cross-analysis cross-tabulations, time series, correlation heatmap, scatter plots; sortable/filterable data table; copy-to-clipboard functionality for all charts; MCP server mode (csvglow --mcp); OpenClaw skill support; smart sampling for large files (100k+ rows); multi-worksheet Excel support; multi-file support with join keys; custom color palettes.
Limitations: Excel currently supports only the first worksheet, and multi-worksheet support is not yet available. Other advanced features such as multi-file joins and custom color palettes are planned but may not yet be available in the current version; the actual project release shall prevail.
