# mcp.browserstack.com agent skills for Claude Code, Cursor and RevampKey

> Ready-made mcp.browserstack.com playbooks for AI agents: sweep BrowserStack accessibility issues into a prioritised fix list, audit your BrowserStack test case library with AI, triage failed BrowserStack builds with AI. Free SKILL.md files for Claude Code, Cursor and Codex, or one click in RevampKey.

## Skills

- [Use BrowserStack from your AI tool, safely](https://revampkey.com/marketplace/mcp-browserstack-com/skills/essentials.md): Confirms your BrowserStack account and project first, routes each request to the right product, and asks for your approval before anything is created or run.
- [Sweep BrowserStack accessibility issues into a prioritised fix list](https://revampkey.com/marketplace/mcp-browserstack-com/skills/accessibility-issue-sweep.md): Reviews the accessibility issues BrowserStack has found on a site or build, groups them by rule and component, ranks them by user impact and drafts a fix list for developers. It reads scan and auth configuration only and never changes them without your approval.
- [Audit your BrowserStack test case library with AI](https://revampkey.com/marketplace/mcp-browserstack-com/skills/test-case-library-audit.md): Walks your BrowserStack Test Management folders and cases to find duplicates, coverage gaps against test plans, and stale or unowned cases, then proposes a cleanup plan. It changes nothing until you approve.
- [Triage failed BrowserStack builds with AI](https://revampkey.com/marketplace/mcp-browserstack-com/skills/test-failure-triage.md): Sorts the failures in a red BrowserStack automated build into real product bugs, test defects, flaky tests and environment problems, using logs, screenshots and root-cause analysis. It reads the evidence and does not rerun or edit anything without asking.
- [Prepare a BrowserStack test run with AI](https://revampkey.com/marketplace/mcp-browserstack-com/skills/test-run-preparation.md): Checks the release test plan, its sub-plans and folders, and drafts the test run scope for you to review. It asks for your explicit yes before creating the run or recording any results.

## Install

- RevampKey: connect mcp.browserstack.com at https://revampkey.com/marketplace/mcp-browserstack-com; the skills come with it.
- Any Agent Skills tool, into the current project:

```bash
for s in essentials accessibility-issue-sweep test-case-library-audit test-failure-triage test-run-preparation; do
  curl -fsSL --create-dirs -o .agents/skills/mcp-browserstack-com-$s/SKILL.md https://revampkey.com/marketplace/mcp-browserstack-com/skills/$s/SKILL.md
done
```

### What is a mcp.browserstack.com agent skill?

A SKILL.md file: a short playbook an AI agent loads when your request matches it. Each mcp.browserstack.com skill describes one job, which mcp.browserstack.com tools to use in which order, what to check first and when to stop and ask you.

### Do the mcp.browserstack.com skills change anything on their own?

2 of the 5 only read. The others tell the agent to show you the change and wait for a yes before it creates, edits or deletes anything in mcp.browserstack.com.

### Which tools can use these skills?

Any agent that reads the open Agent Skills format: Claude Code, Cursor, Codex, VS Code with Copilot and RevampKey. In RevampKey they are attached when you connect mcp.browserstack.com; elsewhere you copy the folders into your project or skills directory.

### Do I need the mcp.browserstack.com MCP server too?

Yes. The skills tell the agent how to do the work; the mcp.browserstack.com MCP server gives it the tools to do it. The mcp.browserstack.com MCP setup guide covers every client.

MCP setup guide: https://revampkey.com/marketplace/mcp-browserstack-com.md