The CheckOps Engineer home

Tools directory

M

The CheckOps Engineer

MonkeyTest

MonkeyTest agents use your site like a real user to find bugs, report them, and verify them once fixed.

Visit website

Latest public signals

Page copy revised 25 Oct 2026 from public sources. Refresh checks the live site and, when this entry has a GitHub repository, the latest release.

Visit Website

Introduction

An AI agent that uses your site like a real user — and reports the bugs. Open source. Playwright + your favourite LLM. First result in about a minute, no API key.

The latest GitHub release listed on 25 Oct 2026 was v0.4.1, published 2026-10-24.

Features

These items are quoted from public documentation fetched on 25 Oct 2026. Taken from repository README headings.

  1. Docker: A multi-stage image compiles this package and leaves it at /opt/monkeytest-core (dist/, package.json, and production dependencies) so another image can copy it, or so a sibling checkout can satisfy file:../monkeytest-core. API keys and Playwright browsers are not in the image. Build, the mtaiserver layout, and the…
  2. Why MonkeyTest Core?: - Closed loop, not just discovery. Stable JSON plans + bug fingerprints + a real rerun and diff — built in, not bolted on. - Drop-in, not framework. One library, one constructor. No DSL. - Bring your own LLM. OpenAI, Gemini, Anthropic, Mistral, Groq, or any OpenAI-compatible endpoint (Ollama, vLLM, OpenRouter). Pick a…
  3. Where it shines: - Pre-release sanity checks on staging. - PR previews — run on Vercel/Netlify preview URLs in CI (recipe). - Daily smoke runs against prod with cheap models. - Embedding in your own product — the engine is a library, not a CLI.
  4. What it isn’t: - Not a replacement for deterministic critical-path E2E tests — it’s a complement. - Not a load tester or a11y auditor. - Not magic — quality of bug reports tracks the quality of the LLM you give it.
  5. Documentation: - Quick start — npm install, or a clone for the no-key smoke run, then Gemini or Ollama - Why monkeytest — how it compares with nearby tools - The bug-finding loop — explore once, run repeatedly, diff - CLI reference — every subcommand and flag - How it works — the four phases - LLM providers — model picks per…
  6. MonkeyTest Cloud: Don’t want to run it yourself? MonkeyTest Cloud is the same engine, fully managed: hosted browsers, history, dashboards, team accounts.

Pros

Points stated in the public sources fetched on 25 Oct 2026:

  1. Published license: The GitHub repository lists GNU Affero General Public License v3.0 (AGPL-3.0).
  2. Closed loop, not just discovery: Stable JSON plans + bug fingerprints + a real rerun and diff — built in, not bolted on.
  3. Drop-in, not framework: One library, one constructor. No DSL.
  4. Bring your own LLM: OpenAI, Gemini, Anthropic, Mistral, Groq, or any OpenAI-compatible endpoint (Ollama, vLLM, OpenRouter). Pick a different model per phase.
  5. Real evidence: Per-step screenshots, optional WebM video, console + network failures, and bug reports with stable fingerprints across runs.
  6. Live activity stream: Subscribe to a typed EventBus and pipe explorer / planner / runner / triager events into your own dashboard.

Cons

The pages fetched on 25 Oct 2026 do not publish a limitations list. That absence is not evidence that the tool has no drawbacks. Current issues are tracked at https://github.com/monkeytestai/monkeytest-core/issues.

The repository README includes this example (from “Install”):

npm install @monkeytestai/monkeytest-core
npx playwright install chromium

Resources:

Pricing

A pricing link was found (pricing page), but the HTML fetched on 25 Oct 2026 did not include a readable price. The page may fill prices in with JavaScript. This rewrite does not guess a number.

The fetched pages do not name a target audience. Decide fit from the official description and the feature list above.

Conclusion

MonkeyTest is summarized by its public description as: “MonkeyTest agents use your site like a real user to find bugs, report them, and verify them once fixed.”.

The latest GitHub release listed was v0.4.1, published 2026-10-24.

This page was rebuilt on 25 Oct 2026 from monkeytest.ai, github.com. It keeps only what those pages stated. Confirm the current product, license, and price on the official site before you adopt it.