> ## Documentation Index
> Fetch the complete documentation index at: https://docs.verisynth.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# CLI Commands

> Command reference for VeriSynth CLI — all supported commands and subcommands.

# CLI Commands

The **VeriSynth CLI** is a lightweight, Python-based command-line interface for generating synthetic datasets and producing verifiable proof receipts.

All commands run **offline** and are designed to integrate easily with Python, CI/CD pipelines, and local analytics workflows.

***

## Base Command: `verisynth`

The base command generates synthetic data and outputs both:

* A **synthetic dataset** (`synthetic.csv`)
* A **proof receipt** (`proof.json`)

```bash theme={null}
verisynth <input_csv> -o <output_dir> [options]
```

Example:

```bash theme={null}
verisynth data/patients.csv -o out/ --rows 1000000 --seed 42
```

Output:

```
📁 out/synthetic.csv
🧾 out/proof.json
```

Result summary:

```
VeriSynth — Synthetic Data Report
========================================
Input: data/patients.csv
Output: out/synthetic.csv
Engine: GaussianCopula | Seed: 42
Fidelity: corr Δ=0.23 | Privacy risk=0.0
Proof: out/proof.json (Merkle verified)
```

***

## Global Options

| Flag       | Description                                          |
| ---------- | ---------------------------------------------------- |
| `--input`  | Path to input CSV file                               |
| `--output` | Output directory for synthetic data and proof        |
| `--rows`   | Number of synthetic rows to generate (default: 1000) |
| `--seed`   | Random seed for deterministic reproducibility        |
| `--schema` | Path to YAML schema configuration file (optional)    |

***

## Schema Configuration

VeriSynth supports explicit field mapping and exclusion through YAML schema configuration files. This gives you fine-grained control over which fields to synthesize and how to handle data types.

### Schema Configuration Format

```yaml theme={null}
exclude: ["patient_id", "address"]
types:
  age: int
  bmi: float
  smoker: bool
  hba1c: float
model:
  engine: GaussianCopula
  seed: 42
```

### Configuration Options

* **`exclude`**: List of field names to exclude from synthesis (e.g., IDs, addresses)
* **`types`**: Explicit type mappings for fields (supports: `int`, `float`, `bool`, `str`)
* **`model`**: Model configuration including engine and seed

### Benefits

* **Privacy**: Exclude sensitive identifiers and PII
* **Control**: Explicit type handling instead of automatic detection
* **Reproducibility**: Schema configuration is included in proof receipts
* **Validation**: Built-in validation ensures configuration correctness

***

## Example Workflows

### Generate 1M rows from a healthcare dataset

```bash theme={null}
verisynth data/patients.csv -o out/ --rows 1000000 --seed 42
```

### Disable proof generation (for benchmarking only)

```bash theme={null}
verisynth data/patients.csv -o out/ --rows 50000 --no-proof
```

### Use a different model (future feature)

```bash theme={null}
verisynth data/transactions.csv -o out/ --model ctgan --rows 250000
```

***