playwright

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill's CLI-first patterns (e.g., "$PWCLI" fill e2 "password123") demonstrate and require embedding credentials or form secrets as literal command arguments, so an agent would likely need to include user-provided secrets verbatim in generated commands, creating an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md and references (e.g., "Quick start", "Core workflow", and references/cli.md) explicitly instruct running pwcli open and using commands like snapshot, eval, click, and fill against arbitrary web pages (e.g., open https://playwright.dev or example.com), so the agent fetches and interprets untrusted public web content that can directly influence subsequent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The wrapper script invokes npx --package @playwright/cli playwright-cli at runtime, which fetches and executes remote code from the @playwright/cli npm package (https://www.npmjs.com/package/@playwright/cli), so the skill relies on executing externally fetched code during runtime.