Skip to content
Get packs

Browser Test Runner

Runs browser tests on the pages a branch or PR changed and reports pass, fail, or skip per route.

Source Download skill
Repository
EveryInc/compound-engineering-plugin
Stars
25,850
Licence
MIT
Files
5 files

What it does

When installed, the agent can look at the files changed on your current branch or a pull request, work out which web pages those changes affect, and open your local dev server in a browser to click through them. It then reports each affected page as Pass, Fail, or Skip, with a reason for any skip. It prefers a browser built into the agent's host app and falls back to a tool called agent-browser, and it is designed not to swap in other browser automation stacks.

Great for

  • Web developers who want a quick browser check of pages touched by their current change before opening a PR
  • Teams running unattended pipelines that should test changed pages and log results without stopping for questions

Skip it if

  • Projects without a web front end or local dev server, since there are no pages to open
  • Flows that need real OAuth, email, payment, or SMS verification, which the skill logs as Skip or pauses for a human to handle

Try asking

  • “Run browser tests for the current branch”
  • “Check browser tests for PR 123”
  • “Run browser tests on branch feature/login using port 4000”
Step-by-step guide: how to use Browser Test RunnerInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 5 original files, the MIT licence and an attribution note.

    ce-test-browser.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip ce-test-browser.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip ce-test-browser.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Start your dev server before manual runs, since the skill verifies it is running and will not start it for you in manual mode
  • Pass --port N explicitly if your project uses a non-default port, rather than relying on detection
  • Review the summary for Skip entries, which mark routes that were not actually exercised

What's inside

  • references/agent-browser-driver.md
  • references/pipeline-orchestration.md
  • references/route-and-report.md
  • scripts/resolve-port.sh
  • SKILL.md

Redistributed unmodified from EveryInc/compound-engineering-plugin under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.