| name | browser-automation |
| description | Optional browser automation workflow for interactive browser tasks, Playwright-style automation, and stable scraping/testing patterns. |
| risk | safe |
| source | codex-max |
| origin | Antigravity-adapted |
| date_added | 2026-04-19 |
Browser Automation
Overview
Use browser-automation for browser-driven tasks that are more interactive or automation-oriented than the normal E2E test workflow.
This includes repeatable browser actions, scraping-style flows, UI automation, popup handling, and resilient selector/wait strategies.
This skill is installed as an optional library skill, not a core routing default.
Use when
- Automating a browser workflow end-to-end
- Building or fixing Playwright-style browser automation
- Handling popups, downloads, or multi-tab flows
- Extracting information from a web UI with stable automation patterns
- Debugging flaky browser interactions caused by selectors or waits
Core principles
- Prefer user-facing locators over brittle CSS/XPath selectors
- Let the automation framework auto-wait when possible
- Avoid arbitrary sleeps and
waitForTimeout-style habits
- Keep browser sessions isolated when test or workflow independence matters
- Collect artifacts when failures are hard to reproduce
Recommended workflow
1. Define the browser goal
State clearly:
- entry URL or app state
- actions to perform
- expected visible result
- whether the task is testing, automation, or extraction
2. Choose stable selectors
Prefer:
- accessible roles
- visible labels
- test IDs
- explicit text anchors when appropriate
Avoid defaulting to brittle DOM-shape selectors.
3. Use condition-based waits
Wait for:
- visible elements
- expected responses
- navigation completion
- popup readiness
- download events
Avoid arbitrary time delays unless no better synchronization point exists.
4. Capture evidence
When needed, retain:
- screenshots
- traces
- videos
- downloaded files
- extracted data snapshots
Output contract
When this skill is used, the result should include:
- Browser goal
- Automation strategy
- Selector/waiting approach
- Failure-risk notes
- Artifacts/evidence strategy
Related skills
e2e-testing — for formal user-journey verificationdocumentation-lookup — for current browser/tooling API usagesearch-first — before adding a browser automation dependency or wrapper
