position-me

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

• Third-party content exposure detected (high risk: 0.95). The required workflow performs a runtime exhaustive crawl of the user-supplied website via browser automation (chrome-devtools/Playwright), which will ingest outsider-authored free text from public pages (e.g., homepage/about/pricing/blog content) into the agent’s LLM context for analysis.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires at runtime connecting to external browser-control components hosted at https://github.com/ChromeDevTools/chrome-devtools-mcp and https://github.com/pasky/chrome-cdp-skill (with a fallback to Playwright), which are external repositories relied upon to perform remote browser automation and thus enable execution of remote code/control during the skill run.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (low risk: 0.30). The skill requires launching/connecting to a local browser with remote debugging enabled, using external devtools/automation tooling and reading local reference files—actions that change and expose runtime state (open debug ports, spawn browser processes, write screenshots)—but it does not ask for sudo, editing protected system files, or creating user accounts, so the risk is present but limited.