Top 200 ReposEXPLORER
Browse agents, categories and pages
Back to results
Reported by sourceThe source says this happened. Top 200 Repos has not reproduced it.

Let an assistant test QA tickets in a browser via Playwright MCP

Peter Eysermans connected Claude Desktop to Playwright and GitHub MCP servers so it could walk through features from tickets, report results and draft reusable tests.

Try the promptRead the original (opens in a new tab)

The problem

How it works

He added the Playwright MCP server and the GitHub MCP server to Claude Desktop. He first had Claude log in to his team's app and test a ticketed feature (creating an event and assigning an employee), then had it fetch GitHub issues labeled qa, follow each issue's embedded testing prompt and comment with the results.

Results and limitations

He reports Claude completed the event form and worked out the required fields itself, got stuck after misreading a role component (which he took as a possible usability flaw), could turn its steps into a reusable Playwright test, and posted issue comments, although a screenshot upload failed.

Before you try it

No counts of tests run, bugs found or time saved, and the session used up his full daily token allowance. He treats it as a first pass that still needs human review, and the GitHub MCP required a classic access token.

Tools and repositories

Tools
Claude Desktop · Playwright MCP · GitHub MCP · Chromium
Prerequisites
Needs Node.js for the Playwright MCP server and Docker plus a GitHub token for the GitHub MCP server.

Try it yourself: Walk through a QA ticket in a browser and report

Fill the fields to adapt the prompt. Empty fields stay as visible [placeholders].

Preview
Using the browser tools, test the feature described in ticket [Ticket ID] on [App URL]. Sign in only with the test account from [Where test credentials live]; never use real user accounts. First restate the steps and expected results you will check, then perform them, noting anything confusing in the interface as a possible usability issue. If you get stuck on the same step twice, stop and describe what you saw. Report each step as pass, fail or blocked with a short note. Then draft a Playwright test file at [Test file to create] covering those steps. Ask before posting any comment to [Issue tracker] or submitting forms that create real data.

Point it only at a local or staging build with test accounts, since it will click buttons and submit forms.

Sources

  1. Claude as tester using Playwright and GitHub MCP (opens in a new tab)

    First-person account · madewithlove.com · Published 22 Jul 2025 · Source checked 9 Nov 2026, 22:20 UTC

    What this source supports
    • Claude Desktop with Playwright MCP and GitHub MCP
    • Tested a ticketed feature and later qa-labeled GitHub issues
    • Got stuck after misreading a UI component
    • Generated a reusable Playwright test; screenshot upload failed
    • Used the whole daily token allowance
Report a problem

Reported by source

Fill out repetitive web forms with Skyvern's vision-based browser agent

Skyvern drives a browser with an LLM and computer vision to complete forms, such as quote requests or contact forms, on sites it has not seen before.

github.commicrosoft/playwright
Reported by source

Prepare a tax return with an agent and compare it to an accountant

Kyle Corbitt worked through his 2025 US taxes with Codex, which researched rules, gathered documents and wrote a custom tax engine while a paid accountant did the same return.

corbt.com · 7 Mar 2026openai/codex, microsoft/playwright
Reported by source

Run parallel Devin sessions on repetitive migrations and test backlogs

Cognition reports customers giving Devin written migration and testing playbooks, then running many sessions in parallel; its own numbers show where Devin does well and where it needs a human.

cognition.ai · 14 Nov 2025
Planned, not done

Watch competitor pricing and product pages with changedetection.io

changedetection.io checks chosen parts of web pages on a schedule, tracks prices and restocks, and can use an LLM to summarise changes or ignore ones that do not matter.

github.commicrosoft/playwright
Reported by source

Compare products across open tabs with Claude in Chrome, then buy yourself

Claude in Chrome can read and compare what is open in your tabs and pull details into a table; Anthropic’s own red-team numbers explain why it asks before purchases and why you should finish checkout.

claude.com · 25 Aug 2025
Reported by source

Give Claude your Google Calendar and Gmail through a self-run MCP server

The open-source Google Workspace MCP server exposes Calendar, Gmail and other Workspace tools to Claude Desktop, Claude Code and other MCP clients, using your own OAuth client and optional read-only mode.

github.com

More with Claude Desktop

Reported by source

Program a robot dog with an AI assistant: Anthropic's Project Fetch experiment

Anthropic split eight staff without robotics experience into two teams to program a quadruped robot to fetch a ball; the team using Claude finished more tasks and, on shared tasks, took about half the time.

anthropic.com · 12 Nov 2025

View all: Claude Desktop