How do I install the Selenium MCP E2e skill?

Run npx skillmds@latest add phungxuananh/selenium-mcp-e2e in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Selenium MCP E2e safe to use?

SkillMD's automated safety review verdict for this skill is PASS. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Selenium MCP E2e?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Selenium MCP E2e free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Selenium MCP E2e?

PhungXuanAnh (@phungxuananh) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Selenium MCP E2e by phungxuananh — Agent Skill

Name: Selenium MCP E2e
Rating: 4.2 (5 reviews)
Author: PhungXuanAnh

Selenium MCP E2e

Run and diagnose this repository's real MCP stdio end-to-end test after any file update. Use before handing off every change to selenium-mcp-server; unit tests, direct Python tool calls, and MCP Inspector checks are not substitutes.

PhungXuanAnh Updated Sep 10, 2026

File contents

Selenium MCP E2E

Validate every repository update through the actual MCP client/server boundary and real Chrome.

Required check

Run from the repository root after the final edit:

.venv/bin/python tests/mcp_stdio_e2e.py -v

The test must exercise all of these layers in one run:

spawn python -m mcp_server_selenium over stdio;
initialize a real MCP ClientSession and discover the affected tools;
invoke the tools through MCP serialization and dispatch;
operate real Chrome through the normal Selenium driver;
verify observable filesystem/browser results;
close the client/server and browser cleanly.

The test owns a temporary user_data_dir. Do not point it at the persistent profile /home/xuananh/.config/google-chrome-selenium-mcp-direct.

Failure handling

If E2E fails, keep the exact failing output, diagnose the server/Chrome state, fix the cause, and rerun the same E2E command. Do not replace it with a unit test, direct tool import, or a cheaper smoke check.

If Chrome or another external dependency genuinely prevents execution, report the blocker explicitly. Do not describe the update as verified.

After the run, confirm no test process remains:

ps -eo pid,ppid,args | rg '([m]cp_server_selenium|[c]hrome).*?/tmp/[^ ]+/chrome-profile' || true

No output is the expected result. Do not stop the persistent MCP server using /home/xuananh/.config/google-chrome-selenium-mcp-direct. Report the E2E command, pass/fail result, and leak-check result in the handoff.

Frequently asked questions