Browser MCP vs Playwright MCP: Reddit debate on when agents need browser tools at all
TwoBlueFools · reddit · 2026-11-07
A Reddit discussion questions whether browser-automation tasks need an MCP at all. The author argues that when a model genuinely explores pages, inspects state, decides what to click, and recovers from surprises, a rich browser/Playwright MCP surface makes sense. But for simple jobs—open a feature, run a flow, confirm it worked—handing the agent a giant bag of browser tools is excessive. That's why Kane CLI appealed to them: a narrow verification command callable from coding workflows, with structured output and exit status, instead of turning the browser into another interactive tool surface. The author isn't claiming it replaces Playwright MCP, but asks where the line sits between "agent needs browser tools" and "agent just needs one boring command that proves the thing works."
More from coding & agent
- Solo founder claims $20K months running an entire team's work through Claude — PrajwalTomar_ · 2026-11-07
- Thoughtworks Technology Radar tabs mattpocock/skills as Trial for agentic coding — mattpocockuk · 2026-11-07
- OnCall-Memory: open-source incident response agent that learns from past production incidents — NewAlarm8213 · 2026-11-07
- Should MCP server operators trust agent identities issued by a third-party service? — zach-benson · 2026-11-07
- Build a 'strategist' agent: route every agent's accumulated context into key business decisions — EXM7777 · 2026-11-07
- Terrarium open-sources local VM sandbox giving AI agents explicit, auditable permissions — v_0009 · 2026-11-07