Playwright native WebMCP testing bridges agent APIs.
Adds page.webmcp and frame.webmcp APIs to test browser-registered WebMCP tools directly.
Brings experimental browser agent tools into standard end-to-end testing frameworks.
Proxies cloud usage APIs, fixes Windows clef head reads, and drops upstreamed Metal residency patches.
Improves local model execution stability and Windows path handling.
Sets GPT-6.1 Sol as the default catalog model and adds active-session MCP server login.
Streamlines multi-agent reasoning defaults and terminal-based MCP authentication.
Testing frameworks are adapting to browser-native agent APIs as WebMCP support lands in Playwright, ensuring teams can automate and verify agent-driven workflows before shipping to production environments; teams should test it in a reversible pilot before adoption.
The same card posted on social. Open it for the full size.
