skills.rest
Cynthia1070711
Cynthia1070711COMMUNITY
Explore all skills by this author →
WebsiteGitHub

autorun-e2e

Launches interactive Claude Code sessions in PowerShell windows to run E2E tests with MCP servers.

10|3|Updated Mar 7, 2026
Search Tags:#claude code#browser automation#e2e testing#pipeline automation#powershell#chrome devtools mcp#auto-chain
One-click install
npx skills add https://github.com/Cynthia1070711/PHYCOOL_Tools --skill autorun-e2e-cynthia1070711
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: autorun-e2e
Source: https://github.com/Cynthia1070711/PHYCOOL_Tools/tree/main/config-templates/claude/skills/autorun-e2e
Command: npx skills add https://github.com/Cynthia1070711/PHYCOOL_Tools --skill autorun-e2e-cynthia1070711

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Running E2E tests through Claude Code requires full interactive mode so MCP servers like Chrome DevTools load correctly, but the standard headless -p pipeline mode cannot do this. This Skill bridges that gap by spawning dedicated interactive PowerShell windows for E2E tasks while keeping pipeline tracking, status updates, and auto-chaining intact. ## Core Features & Use Cases - Interactive E2E Task Windows: Launches full interactive Claude Code sessions (not -p mode) so Chrome DevTools MCP and other MCP servers load for browser automation, screenshots, and DOM verification. - Pipeline Tracking & Watchdog: Writes tracking JSON entries, monitors a signal file or process exit with a 30-minute timeout, and force-kills stalled windows. - Auto-Chain Execution: After E2E completes, reads the Story status and automatically chains to the autorun /dev stage if the Story is not Done, up to a configurable max chain depth. - Use Case: A Story requiring e2e-testing is dispatched by the /fw orchestrator; this Skill opens a window where Claude uses Chrome DevTools MCP to navigate pages, take screenshots, verify UI behavior, and update the Story file before handing off to the next pipeline stage. ## Quick Start Ask the orchestrator to dispatch an E2E test for a Story, or run the pipeline-e2e.ps1 script with -Launch, -StoryId, -Command "/e2e-test", and -StoryPath to open an interactive E2E testing window.

Frequently Asked Questions about autorun-e2e

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run E2E tests with Claude Code and Chrome DevTools MCP?▼

Launch Claude Code in full interactive mode rather than -p headless mode, because MCP servers like Chrome DevTools only load in interactive sessions. This Skill's pipeline-e2e.ps1 script opens such a window with the Story prompt preloaded.

What is the difference between autorun and autorun-e2e?▼

autorun uses headless `claude -p` mode without MCP loading and auto-closes windows, suiting /dev and /code-review stages. autorun-e2e uses full interactive mode with MCP support and manual window closing, suiting /e2e-test tasks.

Why does the E2E window not close automatically after testing?▼

The window is designed for manual closure because interactive sessions may need user intervention for permission prompts or MCP operations. You must exit Claude with /exit so tracking updates and auto-chain logic execute before the window closes.

What happens if an E2E test window hangs or times out?▼

A watchdog polls the signal file and process state every 10 seconds. After the 1800-second timeout it force-kills the window process tree with taskkill, and stale checkpoints older than 30 minutes should be treated as crashed.

Does the E2E pipeline continue automatically after tests finish?▼

Yes. After Claude exits, the script reads the Story status and, if it is not Done, auto-chains to the autorun skill's /dev stage with incremented chain depth, repeating until Done or the max chain depth of 4 is reached.

Related Skills

Explore More Software Engineering Skills→
WizarckWizarck
2

bmad-agent-dev

Executes approved development stories with test-driven implementation and acceptance criteria tracking.

Community
Intermediate
SynkraAISynkraAI
3.1k

aiox-review-story

Reviews a development story and writes a QA gate verdict with lifecycle status transitions.

Official
Intermediate
linenoizelinenoize
1

context-lifecycle

Orchestrate context checkpoints across session boundaries and compaction events.

Community
Advanced
singhianand23045singhianand23045

setup-browser-cookies

Import Chromium cookies into headless browser sessions for authenticated QA testing.

Community
Intermediate