Important: This documentation covers Yarn 1 (Classic).
For Yarn 2+ docs and migration guide, see yarnpkg.com.

Package detail

qlood-cli

jaysym4kMIT0.1.45

Open-source CLI that initializes ./.qlood in your project and uses AI-driven browser automation to test your web app.

cli, testing, web, automation, mcp, openrouter, llm, qa

readme

QLOOD-CLI — Next‑Generation AI Testing Superpower

Supercharge your website and app testing with next‑generation AI automation. QLOOD‑CLI is a preconfigured wrapper around the AugmentCode (Auggie) CLI that thinks like a user, tests like a pro, and catches what humans miss.

Links:

How it works (behind the scenes)

  1. QLOOD‑CLI is pre‑prompted to run specific tests for your app
  2. Orchestrates multiple Auggie CLI runs
  3. Executes tasks with auggie --print "Our prompt"
  4. Launches Playwright (non‑interactive) to test your app and capture screenshots
  5. Auggie drafts results: Errors, Warnings, Success
  6. Auggie analyzes your project and generates a fix prompt automatically

In short: a streamlined auggie --print "Our prompt" wrapper with pre‑made prompts that achieve your goal.

Learn more: https://www.augmentcode.com/

Disclaimer: Using QLOOD‑CLI may require an AI provider API key and an AugmentCode.com account. Running the tool may use your real credits on connected services.

Quick Start

  • Install (global):

    npm install -g qlood-cli

  • Launch the CLI:

    qlood

  • Show all commands (inside the TUI):

    /help

  • Create a workflow (auto‑analyzed and saved to ./.qlood/workflows):

    /wfadd I need a workflow that test the user signup and login

  • List workflows:

    /wfls

  • Run workflow #1:

    /wf 1

  • Run all workflows (great for PR checks):

    /wfall

  • Update workflow #1 based on code changes:

    /wdupdate 1

  • Delete workflow #1:

    /wfdel 1

Results are saved under ./.qlood/results/wf#-%datetime%/ with subfolders:

  • /success — what passed
  • /warning — potential issues (with screenshots) + fix-prompt.md
  • /error — errors (with screenshots) + fix-prompt.md

What QLOOD tests (automatically):

  • End‑to‑end, user‑like flows
  • UI/UX checks and obvious anti‑patterns
  • Console log auditing
  • Network performance signals and failures
  • Security/vulnerability scans (basic heuristics)
  • Auth flow validation
  • Dead link detection and navigation problems
  • API key exposure hints and URL rewrite safety

Requirements

  • Node.js 18+
  • Auggie CLI installed and authenticated: auggie --login

Notes

  • QLOOD initializes a project‑local ./.qlood/ folder and uses Auggie (via MCP Playwright) to drive a headless browser.
  • You’ll be prompted to initialize on first run. Accepting allows Auggie to index your project for context‑aware testing.

Security

  • Credentials should be provided via environment variables or a gitignored .env file.

Open Source

License

MIT