Skill
Web Ui Test
Register Playwright MCP via UTCP and perform web UI testing and verification. Analyze browser snapshots, click elements, fill forms, and return summarized re...
When to use Web Ui Test
Choose if
You're inside an agent loop and want to drive Playwright MCP through a structured skill that handles registration, snapshotting, clicking, typing, waiting, and console-error checks — without dumping raw snapshot data back into your context. Pick this over calling Playwright MCP tools directly when you want enforced output summarization and a known recovery path for the Chrome profile lock failure mode.
Avoid if
You don't have Node.js + npx available, can't run Chrome with the `mcp-chrome` persistent profile, or you need raw snapshot data rather than summaries (the skill explicitly prohibits returning raw snapshots). Also avoid when concurrent Playwright instances may already hold the `mcp-chrome` profile lock — the README documents this as an immediate-exit failure mode requiring manual recovery.
Risk Flags
- MEDIUM scope README documents a recurring failure mode where a previous Playwright Chrome process holds the `mcp-chrome` profile lock, causing immediate exit; recovery requires killing processes and deleting lock files or using a timestamped profile.
- LOW scope README explicitly prohibits returning raw snapshot data — only summaries are permitted, which limits use cases that need the full DOM tree.
- LOW scope Large pages may exceed practical snapshot size; the README suggests targeted evaluation queries and screenshots as workarounds.
Cost
Type: Unknown
Distribution
- ClawHub
web-ui-test- License
- MIT-0