launch

Launch Code OSS builds with isolated profiles and unique debug ports for Playwright automation and dap-cli debugging.

10|1|Updated Jul 13, 2026
One-click install
npx skills add https://github.com/KevinHuangIsLearning/shortestpath-ide --skill launch-kevinhuangislearning
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: launch
Source: https://github.com/KevinHuangIsLearning/shortestpath-ide/tree/main/.agents/skills/launch
Command: npx skills add https://github.com/KevinHuangIsLearning/shortestpath-ide --skill launch-kevinhuangislearning

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @playwright/cli, and includes scripts (resource) components.

What problem does it solve? Developing and testing VS Code itself requires launching Code OSS from sources with a signed-in profile, driving its UI via Playwright over CDP, and attaching debuggers to the renderer, extension host, main, or agent host processes — all without port conflicts or profile corruption when running multiple instances. ## Core Features & Use Cases - Isolated throwaway profiles: Clones an authenticated user-data-dir into a slim temp copy (excluding caches, workspace storage, and logs) so Copilot/GitHub sign-in persists without mutating the source profile. - Automatic port allocation: Picks free ports for CDP, extension host, main process, and agent host debugging, then prints them as JSON for programmatic pickup. - UI automation helpers: Includes a chat-input focus script and a Monaco paste helper that work around Monaco's native-edit-context ignoring Playwright's fill/type commands. - Use Case: An agent working on VS Code chat features launches an Agents window, attaches @playwright/cli to the CDP port, pastes a prompt into the chat input, and simultaneously attaches dap-cli to the extension host port to hit breakpoints. ## Quick Start Launch a Code OSS dev instance with an isolated authenticated profile and give me the CDP port so I can drive the workbench with Playwright.

Frequently Asked Questions about launch

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

FAQPage Schema
How do I launch VS Code from sources for UI automation?▼

Run the launch.sh script (or launch.ps1 on Windows) from the skill's scripts directory. It clones an authenticated profile, allocates free debug ports, waits until the CDP endpoint responds, and prints a JSON line with the ports and paths to attach @playwright/cli.

How do I type into the VS Code chat input with Playwright?▼

Monaco's native-edit-context ignores Playwright's fill and type commands. Use the monaco-paste.sh helper, which dispatches a synthetic ClipboardEvent paste into the focused chat input, or send per-key press commands as a slower fallback.

Can I run multiple Code OSS instances in parallel for testing?▼

Yes. Each launch picks fresh ports, a fresh temp run directory, and its own shared-data-dir, so instances do not conflict. Give each driver a unique @playwright/cli session name via -s= to keep daemon connections isolated.

Why does the launched Code OSS instance ask me to sign in?▼

The source profile lacks a stored GitHub session. On Windows the session lives in the shared-data-dir rather than the profile, so sign in once by running code.bat directly against the source user-data-dir, then future launches inherit it.

Which debug port should I attach dap-cli to?▼

Use extHostPort for extension host code, mainPort for the Electron main process, agentHostPort for the agent host, and cdpPort for renderer workbench code. The launch JSON reports all four ports per instance.