CLI Reference

Zedda ships a zedda command-line tool with 11 subcommands. It is built with Typer and installed automatically by pip install zedda.

The entry point is defined in pyproject.toml as zedda = "zedda.cli:app". The Typer app has add_completion=False and rich_markup_mode="rich".

# Synopsis

zedda [--help] [--version] <command> [args]

Running zedda with no subcommand prints an ASCII-art banner (in the brand orange #E79C65) and the installed version. --help prints the banner followed by Typer’s standard help.

# Commands

Command	Args	Summary
zedda run	PATH [--ai] [--cols COLS] [--out HTML]	Profile a file and print the EDA report
zedda profile	PATH	Alias of run without options
zedda scan	PATH	Fast silent scan
zedda compare	PATH_A PATH_B	Compare two datasets
zedda fix	PATH	Generate fix suggestions
zedda ml-ready	PATH	Check ML readiness
zedda warnings	PATH	Show all data quality warnings
zedda clean	PATH [-o OUTPUT]	Auto-clean with backup + audit
zedda merge	PATHS... [-o combined.csv]	Merge multiple files
zedda report	PATH [-o report.html]	Export HTML report
zedda ask	PATH QUESTION	Plain-English Q&A
zedda info	PATH	Instant file metadata (no full scan)
zedda version	(none)	Print installed version

# zedda run

zedda run PATH [--ai] [--cols COLS] [--out HTML]

Profiles a file and prints the full EDA report. The most flexible command — supports AI insights, column selection, and HTML export.

Flag	Type	Default	Description
--ai	flag	off	Add LLM insights to the report. Requires ZEDDA_AI_KEY env var.
--cols	string	(all)	Accepted but not yet implemented. Prints a yellow warning and profiles all columns.
--out	path	(none)	Save an HTML report to this path in addition to printing the terminal report.

Example:

zedda run data.csv --ai --out rep.html

# zedda profile

zedda profile PATH

Alias of run without options. Equivalent to zedda run PATH.

# zedda scan

zedda scan PATH

Fast, silent scan. Prints nothing on success; prints errors only on failure. Use this in shell scripts where you want the exit code, not the report.

# zedda compare

zedda compare PATH_A PATH_B

Compare two datasets side-by-side. See Comparing Datasets for the full output description.

# zedda fix

zedda fix PATH

Generate copy-paste pandas fix code, grouped by issue type. See Cleaning & Fixing.

# zedda ml-ready

zedda ml-ready PATH

Scores ML readiness (0–100) and emits a copy-paste fix block. See ML Readiness.

# zedda warnings

zedda warnings PATH

Show all data quality warnings ranked by severity. See Python API: warnings().

# zedda clean

zedda clean PATH [-o OUTPUT]

Flag	Type	Default	Description
-o, --output	path	(input path)	Where to write the cleaned file. Default overwrites the input (after backup).

Auto-clean with backup and audit trail. See Cleaning & Fixing.

# zedda merge

zedda merge PATHS... [-o combined.csv]

Flag	Type	Default	Description
-o, --output	path	combined.csv	Where to write the merged file.

Merge multiple files (≥2 required). Validates schema consistency, detects duplicates across files, flags distribution shifts on common numeric columns. See Python API: merge().

Example:

zedda merge jan.csv feb.csv mar.csv -o q1.csv

# zedda report

zedda report PATH [-o report.html]

Flag	Type	Default	Description
-o, --output	path	report.html	Where to write the HTML report.

Export a self-contained offline HTML report. See HTML Reports.

# zedda ask

zedda ask PATH QUESTION

Plain-English dataset Q&A. Tries offline patterns first; falls back to the LLM if ZEDDA_AI_KEY is set. See AI Q&A.

Example:

zedda ask data.csv "any nulls here?"
zedda ask data.csv "which columns should I drop?"

# zedda info

zedda info PATH

Instant file metadata — name, size, path, and row count. Does not run a full scan — uses a fast line count. Useful for quick sanity checks on large files.

# zedda version

zedda version

Print the installed Zedda version. Reads from zedda.__version__.

# Exit codes

All commands return 0 on success and non-zero on error. Errors are raised as ZeddaError and printed with a traceback unless suppressed.

# Next steps

• Python API — the same functionality as Python functions.
• Guides — deeper walkthroughs of each feature.
• Configuration — environment variables.