agentsclimarketplace

Wsl2 browser setup

Skill ArthurZakirov/AgentDesk/skills/wsl2-browser-setup

Make any machine ready for human and AI-agent work.

Install
npx -y skills add ArthurZakirov/AgentDesk --skill wsl2-browser-setup

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Install and configure browsers for Windows plus WSL2. Use when a user wants native Linux Chrome inside WSL to work for normal browsing or sign-in flows, or when a tool in WSL needs a Windows host Chrome or Edge instance exposed over CDP. Prefer this skill before any tool-specific browser automation setup.

SKILL.md

3.6 KB, 803 tokens by cl100k_base, as published. Nobody here has run it

WSL2 Browser Setup

This skill owns browser setup only.

It does not install or configure agent-browser. Use this skill first to make a browser path work from WSL2, then layer any specific automation client on top.

Choose A Method

Use one of these methods depending on the task:

  • Native WSL Chrome: best for interactive browsing, account sign-in, testing whether WSLg browsing works at all, and any case where the user wants Chrome running inside WSL itself
  • Windows host browser bridge: best when a tool in WSL must control a Windows Chrome or Edge instance over CDP

Method A: Native WSL Chrome

Run:

bash "${CODEX_HOME:-$HOME/.codex}/skills/wsl2-browser-setup/scripts/setup-native-wsl-chrome.sh"

The script:

  • verifies WSL2
  • installs Linux google-chrome if it is missing
  • writes the standard Windows .wslconfig networking settings
  • tells the user when wsl --shutdown is required

After the script says restart is needed, have the human run:

wsl --shutdown

Then reopen WSL and continue with the manual Chrome step.

Manual Chrome Step

Open Linux Chrome:

google-chrome

In Chrome:

  • open chrome://settings/security
  • set Use secure DNS to Off
  • do not leave it at OS default (when available)
  • close all Chrome windows
  • reopen google-chrome
  • open https://example.com
  • open one other normal external HTTPS site
  • open the real sign-in page or app you need

If those pages open and the sign-in page renders normally, treat native WSL Chrome as working.

Method B: Windows Host Browser Bridge

Run:

bash "${CODEX_HOME:-$HOME/.codex}/skills/wsl2-browser-setup/scripts/setup-windows-host-browser-bridge.sh"

The script:

  • verifies WSL2
  • finds Windows Edge or Chrome
  • launches the Windows browser with remote debugging enabled
  • asks Windows for elevation to create the port proxy and firewall rule
  • waits until WSL2 can reach the CDP endpoint

This method is still generic browser setup. It exposes a Windows browser to WSL2 over CDP, but it does not connect any specific client to it.

Safe Smoke Tests

Use these first:

bash "${CODEX_HOME:-$HOME/.codex}/skills/wsl2-browser-setup/scripts/setup-native-wsl-chrome.sh" --check-only
bash "${CODEX_HOME:-$HOME/.codex}/skills/wsl2-browser-setup/scripts/setup-windows-host-browser-bridge.sh" --check-only
google-chrome >/dev/null 2>&1 &

What To Skip First

Do not start here:

  • dbus-x11
  • Ubuntu chromium-browser
  • low-level DNS surgery
  • random browser flags
  • manual Windows firewall edits outside the bridge helper

These were explored and were either unnecessary or only useful after the shorter path had already failed.

Bundled Resources

What ships with it: 10 files

14.3 KB alongside SKILL.md, 6 of them executable

agents/

references/

scripts/

Keep looking

Skills are one crate of 326,970. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.