Playwright MCP Server
MCP server from Microsoft for browser automation and testing workflows backed by Playwright.
Host support
Source boundary
The Microsoft Playwright MCP repository describes an MCP server for browser automation and test workflows.
Sources
| Source | Type | Verified | Citation |
|---|---|---|---|
| Playwright MCP GitHub | github | 2026-09-15 | Official Microsoft repository for Playwright MCP. |
Relationships
| Type | Source | Target | Citation |
|---|---|---|---|
| best_for | playwright-mcp-server | browser-automation-agent | Playwright MCP Server is a browser automation MCP server and is recommended by the browser automation agent scenario metadata. |
| exposes | playwright-mcp-server | browser-tool | Playwright MCP Server exposes browser automation tools for agent-assisted UI workflows. |
Compatibility
| Type | Target | Status | Evidence |
|---|---|---|---|
| server_host | codex | verify_required | Playwright MCP can expose browser automation tools to MCP-capable hosts, but Codex host integration should be verified per workspace. |
| runtime_deployment | local-desktop-runtime | supported | Playwright MCP Server runs browser automation through local process tooling and fits local desktop runtime workflows. |
| server_host | visual-studio-code | supported | The Playwright MCP repository describes availability for GitHub Copilot agent workflows in VS Code. |
How should browser automation agent stacks be bounded?
Browser automation stacks should combine tool reliability with runtime isolation, repeatable test evidence, and explicit review for actions that submit forms, mutate data, or call external systems.
What should a research-agent stack verify before implementation?
Verify retrieval sources, browser automation boundaries, citation handling, and whether the selected framework can keep source provenance visible across multi-step synthesis.