playwright

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's examples include CLI commands that embed a plaintext password (e.g., "$PWCLI" fill e2 "password123"), which forces the agent to produce secret values verbatim in its output.

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

• Potentially malicious external URL detected (high risk: 0.90). The wrapper invokes npx to run the remote package "npx --yes --package @playwright/cli playwright-cli" (and the docs suggest "npm install -g @playwright/cli@latest"), which fetches and executes code from the npm registry at runtime, so this is a runtime external dependency that executes remote code.