seo-audit

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

• Third-party content exposure detected (high risk: 0.85). The skill runs seomator audit <url> --format llm (and optionally --crawl), which fetches and parses the target website’s HTML/text at runtime and then feeds that extracted page content into the LLM-formatted output context—i.e., outsider-authored free text from arbitrary URLs/pages.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's On Load steps run installation commands that fetch and execute remote packages at runtime — specifically "npm install -g @seomator/seo-audit" (pulling code from the npm registry, e.g. https://registry.npmjs.org/@seomator/seo-audit) and optionally "npx playwright install chromium" (downloading Playwright/Chromium from Playwright's CDN), and the seomator CLI output (--format llm) produced by that remote code is intended to be consumed by the agent, so these runtime fetches execute remote code and directly influence the agent's prompts.