seo-audit-full

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

• Third-party content exposure detected (high risk: 1.00). The skill's mandatory workflow requires fetching and analyzing live HTML from arbitrary public page URLs (see Phase 1 "fetch-page.py" and Phase 2 "check-social.py" in SKILL.md), so the agent ingests untrusted third‑party content that directly feeds LLM checks and influences audit decisions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill's runtime scripts fetch and parse a user-supplied page URL (example shown as https://example.com) — the fetched HTML is required for the audit and is injected into the analysis/LLM review flow, meaning remote page content can directly influence model prompts/output.