Atlas › 01 Agent Skills for Browser Automation › Token Economics: Why Skills Win on Cost☰ Read as one page
Token Economics: Why Skills Win on Cost
3.1The Context Window is a Shared ResourceEvery token in the context window competes for the same limited space. A coding agent typically needs context for:3.2MCP Token Cost: The Tool Schema TaxEvery MCP tool exposes a JSON schema that's loaded into every API request:3.3Skill Token Cost: The Markdown InjectionA browser-automation SKILL.md (Playwright CLI's generated skill, or Vibium's vibe-check before it) is ~100 lines, which translates to…3.4Side-by-Side ComparisonNote the gap between the naive per-step arithmetic and the measured end-to-end number, because it's an architect-level insight: in a real…3.5Why This Matters for Test AutomationA typical test automation session involves: - Reading test specifications - Writing test code - Running browser interactions - Analyzing…3.6Microsoft Settled the ArgumentIn early 2026 Microsoft shipped @playwright/cli — a CLI purpose-built for coding agents, with a --skills flag that generates a SKILL.md…3.7When MCP Token Cost Is JustifiedMCP's higher token cost buys you:3.8Interview Talking Point"When we evaluated browser automation approaches, token economics was the deciding factor. Our test sessions run dozens of browser commands…