browser-testing-with-devtools

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

• Third-party content exposure detected (high risk: 0.75). Outsider-authored free text can enter the LLM context via the Chrome DevTools MCP tools that read browser content at runtime—specifically DOM text, console logs, and network response bodies (all treated as untrusted data) which the agent may include in its reasoning context.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's runtime configuration runs "npx -y chrome-devtools-mcp@latest --isolated" (fetching and executing the chrome-devtools-mcp package from the npm registry) and the skill requires this external package to operate, so it downloads and executes remote code at runtime.