seo-audit

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly fetches and renders arbitrary public website content via Bright Data (e.g., the workflow and recipes instruct bdata scrape <url> -f html in SKILL.md and references/bdata-recipes.md R-04/R-07 and runs bdata search "site:<domain>" in R-12), then parses and acts on that untrusted, user/third-party-generated content to produce findings and drive follow-up checks, so third-party pages could inject instructions that materially influence the agent's decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The SKILL explicitly requires running a remote installer via curl -fsSL https://cli.brightdata.com/install.sh | bash to install the Bright Data CLI (a required dependency) which would fetch and execute remote code on the host, so this external URL poses a high runtime code-execution risk.