Skip to content

Command: auto

The auto command executes the full autonomous testing lifecycle in a single atomic operation. It is the primary command for end-to-end verification and CI/CD integration.

Technical Execution Flow

When invoked, the auto command orchestrates the following internal phases:

  1. Source Discovery: Recursively scans the target directory for supported source files.
  2. Context Analysis: Extracts AST metadata and identifies code blocks requiring tests.
  3. AI Generation: Interfaces with the configured LLM to generate test suites.
  4. Environment Setup: Initializes temporary test directories and captures environment state.
  5. Multi-Language Execution: Triggers native test runners for each identified language.
  6. Result Aggregation: Collects execution data and generates a unified reporting artifact.

Usage Specification

testgen auto [TARGET_DIRECTORY] [OPTIONS]

Arguments

Argument Description Required
TARGET_DIRECTORY The path to the source code root directory for analysis. Yes

Options

Option Shorthand Description Default
--output -o Specifies the directory for generated tests and reports. ./tests
--verbose -v Enables detailed step-by-step execution logs. False
--debug Enables full stack traces and debug logging. False

Execution Example

[bold cyan]TestGen-AI: Autonomous Workflow Initiated[/bold cyan]
--------------------------------------------------
[blue]Phase 1[/blue]: Scanning directory: ./src
[green]Found[/green]: 12 Python files, 4 Javascript files.

[blue]Phase 2[/blue]: Generating test suites...
[green]Generated[/green]: test_auth.py, test_db.py, api.test.js

[blue]Phase 3[/blue]: Executing tests...
[dim]Running pytest...[/dim] ✅ 85% coverage
[dim]Running jest...[/dim]   ✅ 92% coverage

[blue]Phase 4[/blue]: Finalizing Report...
[bold green]Success[/bold green]: Report generated at ./docs/report_2026-02-19.html