# Test Generator

> Generate and run unit, API, and E2E tests (including Playwright) that match project conventions. Use when adding coverage or reproducing a bug with a failing test first.

- Skill: `navinspire-ia/test-generator` (Agent Skill)
- Install (CLI): `npx skillmds@latest add navinspire-ia/test-generator`
- Raw SKILL.md: https://api.skillmd.com/api/skills/navinspire-ia/test-generator/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Integrations & APIs
- Author: Navinspire-ia (https://skillmd.com/u/navinspire-ia)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/navinspire-ia/test-generator

---


# Test Generator

## Overview

Write tests the project already understands (pytest, vitest, jest, playwright). Prefer failing test → fix → green.

## Workflow

1. Detect the test runner and layout (`package.json`, `pyproject.toml`, existing tests).
2. Choose level:
   - unit for pure logic
   - API/integration for handlers
   - E2E for critical user paths
3. Add a focused test next to siblings; mirror naming/fixtures.
4. Run the smallest command that exercises the new test.
5. Iterate until green; avoid flaky sleeps - use explicit waits.

## Rules

- Do not snapshot huge blobs without need.
- Keep tests deterministic; inject clocks/network.
- For UI E2E, reuse `playwright-browser` guidance.