3,873 mods in this category, of every kind an agent can take. Each one carries what it costs per session, what the scan found, and whether it is the original.
Skill Claude CodeCodex
A browser automation testing skill combining Playwright's precise controls with browser-use's AI-guided exploration. It operates a shared Chromium browser session.
Skill Claude CodeCodex
Verify and debug your own web UI on localhost. Use when you changed frontend code and need to check it actually works — console/network errors, layout/contrast/a11y problems, dead buttons, wrong colors, broken handlers — or when testing a dev server, reproducing a UI bug, or debugging why an element behaves wrong.…
MCP server Claude CodeCodexCursor +2
MCP server that runs goal-driven Playwright browser sessions and exposes the full action trace plus aria-snapshots so the calling LLM can verify whether the goal was met. Runs locally from the cowork-qa-mcp npm package. Needs 2 environment variables to run.
hooplus1ce/qa-automation-plugin
Plugin Claude Code
Bundles 2 skills · 89 tokens together
A Chinese-language MCP service for enterprise web testing across SCM, MOM, WMS, and ERP systems. It uses Playwright to control a local Chrome browser and supports page analysis, interactions, action sequences, test recording, and Shadcn Excel export.
mobile-agent-platform/mobile-agent-platform
Skill Claude CodeCodex
A skill for running step-by-step tests on a real Android phone, using screen-reading and device-control tools while checking the result after each action.
Skill Claude CodeCodex
Automated desktop UI testing through the Hyprland MCP server. Captures screenshots, analyzes them via the vision skill, clicks and types, and re-verifies. Use when asked to test a desktop app, verify UI behavior, run a visual regression check, or drive a background app without interrupting the user. Triggers: "test…
Plugin Claude Code
Bundles 1 skill, 1 MCP server · 74 tokens together
Record browser actions as a replayable Trace; deterministic replay with network assertions.
Cursor rule Cursor
LOCKED QAFusionX complete QA flow — 15 steps, n8n-style gates, Ask mode first, visible browser, no skip.
MCP server Claude CodeCodexCursor +2
MCP server "web-ui-tester" as configured in hofmeister/web-ui-tester. Runs locally from the web-ui-tester npm package.
Skill Claude CodeCodex
A proof-based, zero-hallucination UI and API testing agent.
MCP server Claude CodeCodexCursor +2
AI-powered test automation: run tests, analyze failures, query results across mabl workspaces. Remote server at mcp.mabl.com.
MCP server Claude CodeCodexCursor +2
AI agent testing: replay real sessions against a rebuilt staging environment to catch regressions. Remote server at api.runmirrors.com.
Skill Claude CodeCodex
Runs an end-to-end test from a Jira ticket number. Reads the ticket, executes the browser test steps it contains with Playwright, optionally scores any AI or text output with the RetriEval MCP, and writes a pass/fail result back to the ticket. Use this WHENEVER the user gives a ticket key and asks to test, verify, QA…
MCP server Claude CodeCodexCursor +2
Disposable inboxes for AI agents and CI: catch the verification email, read the OTP or link. Runs locally from the @mailsink/mcp npm package. Needs 2 environment variables to run.
hubertgajewski/playwright-report-mcp
MCP server Claude CodeCodexCursor +2
Run Playwright tests and surface structured results for AI agents doing test failure analysis. Runs locally from the playwright-report-mcp npm package. Needs 2 environment variables to run.
inSideos-designs/cowork-qa-mcp
MCP server Claude CodeCodexCursor +2
Goal-driven Playwright browser sessions for LLMs, with full action traces and aria-snapshots. Runs locally from the cowork-qa-mcp npm package. Needs 2 environment variables to run.
Plugin Claude Code
Bundles 7 skills, 2 MCP servers · 666 tokens together
Teach your coding agent to test real mobile apps and websites on the Robot Actions device cloud: connect, drive devices reliably, and record/replay regression flows.
MCP server Claude CodeCodexCursor +2
MCP server "cypress-mcp" as configured in miroslavmyrha/cypress-mcp. Runs locally from the @miroslavmyrha/cypress-mcp npm package.
Skill Claude CodeCodex
Scope coverage, build a testing strategy, test the feature with agent-browser, capture screenshots/video (always mandatory), upload to S3, and create a structured test report (never skip).
Skill Claude CodeCodex
Verifies Kelos skills.sh installation. Use when asked to run the Kelos e2e skill check.
MCP server Claude CodeCodexCursor +2
Enhanced ADB MCP Server for Flutter/Android Development - 60+ tools for UI testing, debugging, and visual QA. Runs locally from the r-adb-mcp-server Python package.
tempestim/ghostinspector-mcp-server
MCP server Claude CodeCodexCursor +2
MCP server "ghostinspector-mcp-server" as configured in tempestim/ghostinspector-mcp-server. Runs locally from the ghostinspector-mcp-server Python package.
Skill Claude CodeCodex
A command-line testing setup for checking bot and plugin commands in PowerShell, Command Prompt, or Bash without connecting to real QQ accounts or channels.
cookiemonster0921/vincent-vnc-mcp
Skill Claude CodeCodex
Structured, reviewable test/automation/demo runs against a remote Mac controlled via the vnc-mac-agent MCP tools (screen/mouse/keyboard/clipboard/apps). Records each step with a screenshot, purpose, feature tested, and a self-assessed evaluation with cited evidence, reviewable afterward at the viewer's /reports page.…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: