claude-marketing-skills

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly ingests and analyzes public, user-generated web content — e.g., BrandDNAExtractor.extract(url="https://example.com") in SKILL.md, content_strategist sources=["rss_feeds","web_scraping"], and agents/competitor-analyst that analyze competitor ads from public ad libraries — so untrusted third-party pages could influence agent decisions and follow-up actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The MCP configuration launches remote npm packages at runtime (e.g., "npx -y @klaviyo/mcp-server-klaviyo" and "npx -y @google-ads/mcp-server" in ~/.claude/mcp_settings.json), which will fetch and execute remote code that the skills rely on, so these are runtime external dependencies that execute remote code.