cesto-toolkit

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to display a returned "loginUrl" value (magic-link) to the user, which can contain a one-time auth token and thus requires outputting a secret-like value verbatim, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflows and scripts (e.g., scripts/fetch_baskets.py, scripts/fetch_basket_detail.py, and scripts/analyze_investment.py) actively fetch and parse public, user-generated content from backend.cesto.co endpoints (GET /products, GET /products/{slug}, /products/analytics, GET /products/{id}/analyze), and that content (basket titles, descriptions, allocation rationales, etc.) is consumed and used to drive recommendations and actions, so it can carry indirect prompt-injection payloads.