Vizb CLI
A CLI that turns CSV, JSON and benchmarks into interactive ECharts charts and statistics in one HTML file, for people and agents
- 91 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The program reads files and API responses you pass to it and writes HTML files, and the API in the Docker image has no authentication.
Install
- Works in
Manual install
npx skills add goptics/skills --skill vizbThe agent skill, after installing the `/vizb` command works.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The program reads files and API responses you pass to it and writes HTML files, and the API in the Docker image has no authentication.
This is third-party code. Review the repository files before installing.
What it does
Vizb is a tabular data visualization engine. It takes CSV, JSON or benchmark output, turns numeric columns into interactive Apache ECharts charts and descriptive statistics and writes everything to one self-contained HTML file: no server, dependencies or build step are needed. The chart type is chosen by a command, for example `vizb bar data.csv -o out.html`. Data is grouped and filtered with flags. `--group` and `--group-pattern` split a field along axes, for example a date into year, month and day for a 3D chart, `--json-path` pulls a nested array out of an API response, `--select` keeps and renames columns, `--stat` adds a statistics panel and `--output` sets the file. The README example builds a 3D chart of GitHub contributions with a single `curl` and `vizb` pipeline. For agents there is a skill: `npx skills add goptics/skills --skill vizb`, after which the `/vizb` command works in Claude Code, Cursor, Codex and other agents. For automation there is a Docker image with an HTTP API, a GitHub Action for CI and installation by script, binary or Docker.
Who it is for. Developers and analysts who need to show benchmark results or a table as an interactive chart quickly, with no server to set up.
Good fit when
- You need a chart from Go benchmark results or other output that opens in a browser and can be shared as a file
- An agent should build a diagram from a CSV or an API response with one `/vizb` command
- Data sits in nested JSON and you need to pull an array with `--json-path` and rename columns
- A date has to be split into year, month and day and shown as a 3D grouped chart
- A performance report has to be built in CI with the GitHub Action and saved as HTML
Not a fit when
- You need a live dashboard with refreshing data: Vizb produces a static HTML file
- You need a complex BI report connected to databases: the tool works with files and streams
- You need a public service with sign-in: the HTTP API in the Docker image has no built-in authentication
Example request
/vizb as bar: take data.csv, group by month, show stats and save the result to report.html.Limitations
The result stays a static HTML file, charts are built from numeric columns and unusual data shapes need tuning of `--group`, `--json-path` and `--select`. The HTTP API in the Docker image has no built-in authentication: keep it on an internal network or put authentication, TLS and access control in front of it. The install scripts run through `curl | bash` and `irm | iex`, so read them before running or take a prebuilt binary from the releases page. Building from source needs Go 1.26. The agent skill relies on the `npx skills` manager.
How to disable. Delete the `vizb` binary and the skill from the agent skills directory. Stop the container with `docker stop` and remove the `goptics/vizb` image. Generated HTML files stay on disk.
FAQ
How do I install Vizb for an agent?
Run `npx skills add goptics/skills --skill vizb`. After that the `/vizb` command works in Grok, Claude Code, Cursor, Codex and other agents the `npx skills` manager supports.
Which data formats are supported?
CSV, JSON and benchmark output. The `--json-path` flag handles nested JSON and `--select` keeps the columns you need.
Is a server needed to view the result?
No. The result is written to one HTML file with charts and statistics that opens in a browser.
How do I run it with Docker?
For the CLI run `docker run --rm -v "$PWD:/data" -w /data goptics/vizb bar data.csv -o out.html`. For the HTTP API run `docker run -d --rm -p 8080:8080 goptics/vizb`, which has no built-in authentication.
Which chart types exist?
The command takes the chart type as its first argument, for example `bar`. The full list is in the Charts section of the documentation.
Discussion
No questions or reviews yet.
Related
CLI toolsBI as code: YAML dashboards, a semantic layer and a local MCP server through which an agent drafts charts.
FlexViz builds dashboards over 100 million rows that stay interactive, and the agent reads back what you selected.
VisActor Skills: charts, dashboards and posters
VisActor Skills
Nine VisActor skills: business charts, data stories, map dashboards and posters, all assembled by your agent
VisiData opens multiple table formats in a terminal and helps explore, arrange and compare data through keyboard controls.
A data visualization skill for agents: turns data into polished interactive HTML charts and full reports with one visual language
A skill that picks a scientific chart from nearly 100 templates for CSV or Excel data and outputs PNG, SVG, PDF and Python code.
README in short
The README describes install options (an agent skill, scripts, a binary, Docker), an example with a GitHub contribution chart, the flags used, a warning about the API without authentication and documentation links.