Skip to content
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 people14/20 fit
TL;DR

Build Bug Report to Repro Script Agent for qa engineers and open-source maintainers in ~30h with 2–3 people. Stack: Playwright + LLM + LangGraph + Node.js.

Why it wins: A failing test is concrete proof the agent worked.

Build flow

01 · Problem
Maintainers waste hours trying to reproduce bugs from unclear reports.
02 · Solution
The agent reads the report, explores the app with a browser, records the steps that trigger the error and outputs a failing Playwright test.
03 · Build
  1. 1Parse the report into hypotheses
  2. 2Drive the app with Playwright tools
  3. 3Detect the error condition
  4. 4Write and run the test
04 · Demo
Feed a real-style report and run the generated failing test.

MVP checklist

  • Report parsing
  • Autonomous exploration
  • Failing test output
  • Step screenshots

Tech stack

TechRole
PlaywrightBrowser automation
LLMLanguage model
LangGraphMulti-agent graph
Node.jsBackend runtime

Reference repos

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

Related ideas

More AI Agents

  • AI Agents

    Dependency Upgrade Agent

    An agent upgrades a dependency, fixes breaking changes and opens a PR with test results

    Advanced28h1–3
    LLMGitHub APINode.js
    15/20

  • AI Agents

    Exploratory QA Agent

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

    Advanced28h1–3
    browser-usePlaywrightLLM
    15/20

  • AI Agents

    Accessibility Fix PR Agent

    An agent finds accessibility issues and opens a PR that fixes them

    Advanced26h1–3
    axe-corePlaywrightLLM
    16/20