Test Tagging and Playwright Project Selection

SkillWeb & browsing

Lets your agent tag E2E tests so they run on the right devices, platforms, and execution contexts.

Use Test Tagging and Playwright Project Selection in Claude, ChatGPT or Ahel Desktop

Free. Sign in, add Test Tagging and Playwright Project Selection and connect your AI. About a minute.

Also: Claude Code · Cursor · Codex

Then ask your AI: use the Test Tagging and Playwright Project Selection skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Test Tagging and Playwright Project SelectionStart free
About this skill

Test tagging and Playwright project selection for Suite E2E tests: device model, platform and execution scope tags. Use when writing or editing tests to control where they run.

What this skill tells your AI

The instructions your AI receives, as published by trezor/trezor-suite in suite/e2e/skills/tagging/SKILL.md and read by ahel’s review.

When to use: When writing new E2E tests or editing old ones to ensure they run on the intended configurations (device models, platforms, execution contexts). Tags are how tests select which projects execute them. Prerequisites: Suite base fixture for device setup and basic E2E test structure knowledge.

Tag Categories

Device Model Tags (Mandatory for device tests)

  • @T3W1, @T3T1, @T3B1, @T2T1, @T1B1 — specific hardware
  • @noDevice — tests requiring no device (API/settings only)

Default targeting strategy: Use ['@T3W1', '@T3T1'] by default unless:

  • Feature doesn't work on one model (exclude that model)
  • Feature is model-specific (target only that model)

Platform Tags (Mutually exclusive)

  • @webOnly — browser environment only; excluded from desktop projects. Required for tests using page.route() (routing is experimental in Electron and doesn't work reliably on desktop).
  • @desktopOnly — Electron app only; excluded from web projects. Required for tests that interact with native OS features (e.g., file dialogs, notifications).
  • (omit both) → runs on all platforms

Execution Context Tags

  • @skipOnPR — never runs on a PR, not even when the LLM test selector or an edited test file targets it. Required for tests that break or mislead on a PR: nonce collisions when two runs overlap, old app versions that only the deployed instances provide. Runs in nightly, canary, and release.
  • @optional — excluded from the full PR run; runs nightly, and on a PR only when the LLM test selector picks it or the PR edits its file. Use for tests that are low priority and slow. When the reason is not obvious from the test itself, state it in a one-line comment above the describe. Combine with a single platform (@webOnly or @desktopOnly) to conserve resources.
  • @specificFirmware — required when using test.use({ firmwareVersion: '2.10' }). Excluded from canary FW projects to prevent mismatched firmware runs.
  • @group=manual — excluded from all automated runs, reserved for manual test definitions

Signals

GitHub stars
1k
Forks
384
Last commit
Sep 2026
Advanced
Item type
skill
Key
tagging
Source
github.com/trezor/trezor-suite