stack-evaluation

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

• Third-party content exposure detected (high risk: 0.90). The skill set (notably cost-aws, cost-gcp, cost-firebase, cost-supabase, and cost-vercel) explicitly instructs the agent to fetch and parse live pricing/docs from public vendor websites (e.g., cost-aws "Step 3: Fallback to Web Search" citing https://aws.amazon.com/pricing/ and the AWS calculator, and similar web/API fallbacks in other cost-* SKILL.md files) and then uses that fetched content to compute cost estimates and drive stack recommendations, which clearly exposes the agent to untrusted third‑party web content that can influence decisions.