Grade A · verified 2026-10-14
Playwright MCP client compatibility
Short answer: grade A: documented for all 5 clients. Based on docs, not hands-on tests.
Transport: stdio · Auth: None · Run / URL: npx @playwright/mcp@latest · server docs ↗ · repo ↗
| Client | Status | Source | Hands-on test |
|---|---|---|---|
| Claude Code | Documented | client MCP docs ↗ | Untested |
| Codex CLI | Documented | client MCP docs ↗ | Untested |
| Cursor | Documented | client MCP docs ↗ | Untested |
| Gemini CLI | Documented | client MCP docs ↗ | Untested |
| VS Code (Copilot) | Documented | client MCP docs ↗ | Untested |
README has setup sections for Claude Code, Codex, Cursor, Gemini CLI and VS Code.
Moving this setup to another agent? Paste your config into the free converter: it outputs the exact files for Claude Code, Codex CLI, Cursor and Gemini CLI.
Badge for your README
[](https://getaicodingagent.com/mcp/playwright)Methodology
- Documented: the server's own README or docs give setup steps for that client (linked on each server page).
- Configurable (derived): the server's docs don't mention the client, but the transport and auth it needs (stdio + env vars, or Streamable HTTP + header/OAuth) are documented by the client's official MCP docs: Claude Code ↗, Codex CLI ↗, Cursor ↗, Gemini CLI ↗, VS Code (Copilot) ↗.
- Grade A: documented for all 5 clients. Grade B: documented for some, derivable for the rest. Grade C: a client is blocked or unsupported (none in this batch).
- Hands-on testing: not yet done. These ratings are based on documentation only, not on running each server in each client. Client allowlists that docs don't mention would not show up here.
- Verified 2026-10-14. Server authors: open an issue to request a re-check.