Vizb

Quick Install

Linux / macOS

curl -fsSL https://vizb.goptics.org/install.sh | bash

Windows

irm https://vizb.goptics.org/install.ps1 | iex

Quick Example

Point Vizb at a CSV. Numeric columns become charts. Categorical columns become axes when you group (or when auto-group picks one for you).

After install, chart a table like this:

sales.csvCopy CSV

region	product	quantity
West	Widget	39
West	Gear	20
West	Sensor	28
West	Relay	14
West	Bolt	33
South	Widget	15
South	Gear	42
South	Sensor	47
South	Relay	22
South	Bolt	18
North	Widget	28
North	Gear	43
North	Sensor	19
North	Relay	36
North	Bolt	25
East	Widget	31
East	Gear	22
East	Sensor	35
East	Relay	12
East	Bolt	40
Central	Widget	35
Central	Gear	17
Central	Sensor	19
Central	Relay	29
Central	Bolt	24

CLI

vizb bar sales.csv -g region,product -p x,y -l -o sales.html

HTTP

POST / on vizb serve. The API takes inlineinput, not a file path.

{
  "input": "region,product,quantity\nWest,Widget,39\nWest,Gear,20\nWest,Sensor,28\nWest,Relay,14\nWest,Bolt,33\nSouth,Widget,15\nSouth,Gear,42\nSouth,Sensor,47\nSouth,Relay,22\nSouth,Bolt,18\nNorth,Widget,28\nNorth,Gear,43\nNorth,Sensor,19\nNorth,Relay,36\nNorth,Bolt,25\nEast,Widget,31\nEast,Gear,22\nEast,Sensor,35\nEast,Relay,12\nEast,Bolt,40\nCentral,Widget,35\nCentral,Gear,17\nCentral,Sensor,19\nCentral,Relay,29\nCentral,Bolt,24\n",
  "parser": "csv",
  "grouping": {
    "columns": [
      "region",
      "product"
    ],
    "pattern": "x,y"
  },
  "charts": {
    "types": [
      "bar"
    ],
    "configs": [
      {
        "type": "bar",
        "showLabels": true
      }
    ]
  },
  "output": {
    "format": "html"
  }
}

Action

GitHub Action step. Same conversion as the CLI.

- uses: goptics/vizb@v0
  with:
    file: sales.csv
    group: region,product
    group-pattern: x,y
    charts: bar
    chart: "bar:labels"
    output-html: sales.html

Open sales.html in a browser. You get a grouped bar chart: region on X, product as series, height = quantity.

Flags in this example:

• bar — one chart type (smallest learning path)
• sales.csv — the table above (use Copy CSV to grab it)
• -g region,product — columns that become dimensions
• -p x,y — first group column → X, second → Y series
• -l — show value labels on bars
• -o sales.html — self-contained HTML

No -g? Vizb auto-groups a high-cardinality categorical column and logs the choice. Override whenever the guess is wrong.

vizb bar sales.csv -o sales.html

From a clone of the Vizb repo, the full sample isexamples/csv/sales.csv (10,000 orders across 2024–2025).

Tip

Advanced demo (date split + remote JSON): the GitHub Legends 3D contribution skylines.

Start from a goal

IntroductionWhat Vizb is and your first chart.

DimensionsName, X, Y, and Z on every chart.

Chart a CSV / JSON tableNumeric columns, grouping, aggregation, and --json-path.

Pick a chart typeHow bar, line, scatter, pie, heatmap, radar, Sankey, and Chord read x/y/z.

Benchmark outputGo, Rust, and JavaScript bench parsers.

CI and merge historyGitHub Action, stateless and stateful tracking.

Capabilities

Tables and benchmarks

CSV and JSON tables, plus Go, Rust, and JavaScript benchmark output. Format is detected from content. See Supported inputs.

Shared dimension model

Name, X, Y, and Z drive every chart type. Group for category breakdowns; select for coordinate plots. See Dimensions.

Self-contained HTML

Interactive UI in one file: sort, swap axes, scale, themes, JPEG export, optional stats. Open it in any browser. See UI.

CLI, Action, and API

Local CLI, composite GitHub Action, and REST for the same pipeline. Merge datasets across runs when you need history.

Full inventory: Features. Live samples: Examples.