Skip to content
AI Agents

Exploratory QA Agent

An agent clicks through your web app like a curious user and reports bugs.

Advanced28h1–3 people15/20 fit
TL;DR

Build Exploratory QA Agent for startups without a qa team in ~28h with 1–3 people. Stack: browser-use + Playwright + LLM + Python.

Why it wins: Finding real bugs live is a great demo.

Build flow

01 · Problem
Small teams ship without testing the paths real users take.
02 · Solution
Using a browser agent, explore the app from a goal list, capture console errors, broken links and confusing states, and write a bug report with screenshots.
03 · Build
  1. 1Define user goals
  2. 2Run browser-use sessions
  3. 3Capture errors and screenshots
  4. 4Summarise findings
04 · Demo
Point it at a demo app and show three real bugs it found.

MVP checklist

  • Goal-driven exploration
  • Console error capture
  • Screenshot bug reports
  • Coverage map

Tech stack

TechRole
browser-useLLM browser automation
PlaywrightBrowser automation
LLMLanguage model
PythonBackend / scripting

Reference repos

Open-source projects you can build on. Links checked 2026-10-31.

Related ideas

More AI Agents

  • AI Agents

    Bug Report to Repro Script Agent

    Paste a vague bug report and the agent tries to reproduce it as a Playwright test

    Advanced30h2–3
    PlaywrightLLMLangGraph
    14/20

  • AI Agents

    Web Form Filling Agent

    Completes repetitive multi-step web forms from a short instruction

    Advanced48h2–4
    PythonPlaywrightLLM
    15/20

  • AI Agents

    Price Drop Watcher Agent

    Tell the agent what you want and your budget and it watches prices for you

    Beginner10h1–2
    browser-usePythonCron
    16/20