9.4Strategy 3: Different Tools for Different Test Types

The Pattern

Test Type	Primary Tool	Why
Functional tests	CLI (playwright-cli)	Known flows, ref-driven, fast
Accessibility audits	CLI snapshot (YAML a11y tree)	The snapshot is an accessibility summary — grep it, reason over it
Visual regression	CLI + screenshot	Files on disk, compared without entering context
Exploratory testing	CLI snapshot + agent reasoning	Discovery is a read away
Sandboxed / hosted agents	Playwright MCP	No filesystem — inline state is the only option
API integration checks	CLI + eval	playwright-cli eval for XHR/state inspection
Performance checks	CLI + eval	playwright-cli eval "JSON.stringify(performance.timing)"

Configuration

# test-framework-config.yaml
test_types:
  functional:
    tool: playwright-cli
    session: per-test          # fresh named session per test
    headed: false

  accessibility:
    tool: playwright-cli
    artifacts: snapshots       # keep YAML snapshots for audit

  visual:
    tool: playwright-cli
    screenshots: true
    comparison_threshold: 0.98

  hosted_exploratory:
    tool: playwright-mcp       # sandboxed surface, no filesystem
    snapshot_mode: full