The Agent Skills Directory

[PROMPT_INJECTION] (LOW): The skill utilizes Playwright for browser automation, creating an attack surface where untrusted data from web pages or console logs could influence agent behavior.
Ingestion points: Untrusted data enters the agent context through page.goto() and page.on('console') in e2e/auth.spec.ts.
Boundary markers: Absent. The snippets do not demonstrate the use of delimiters or 'ignore' instructions when interacting with page content.
Capability inventory: The skill uses pnpm exec playwright (recorded in SKILL.md) which allows network operations, file system access for reports/screenshots, and subprocess execution.
Sanitization: No sanitization or validation of external content (like console error messages) is demonstrated.
[CREDENTIALS_UNSAFE] (SAFE): Hardcoded dummy credentials test@example.com and password123 are used in example code for a local Firebase Emulator.
[COMMAND_EXECUTION] (SAFE): The skill defines standard testing commands using pnpm for running E2E suites.

e2e-playwright