playwright

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill's CLI examples and form-fill commands embed plaintext credentials (e.g., fill e2 "password123"), which implies the agent may ask for and emit real secret values 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: 0.75). The required runtime workflow runs npx --package @playwright/cli playwright-cli (Playwright CLI) which fetches and ingests the contents of the user-specified web pages (e.g., via pwcli open https://...), so any outsider-authored page text can enter the agent’s LLM context through the CLI’s snapshot/DOM-to-text output.

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

• Potentially malicious external URL detected (high risk: 0.90). The wrapper script invokes "npx --yes --package @playwright/cli playwright-cli", which at runtime fetches and executes the @playwright/cli package from the npm registry (remote code execution), and the skill depends on that fetched package to run.