competitor-profiling

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

• Third-party content exposure detected (high risk: 0.85). At runtime, the skill uses Firecrawl to scrape outsider-authored competitor website pages (e.g., homepage/pricing/features/about/customers/integrations/changelog) into readable markdown, which is then placed into the agent’s LLM context for synthesis—this is indirect prompt injection risk from public web content.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly scrapes user-supplied competitor sites at runtime (e.g., https://competitor1.com) via Firecrawl and injects those raw scrapes and DataForSEO responses into the synthesis step, so fetched external content is required at runtime and directly controls the agent's outputs.