recon-cafes

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill includes commands that extract and print API keys (e.g., grep for 'AIza...' ) and shows requests that embed cookies/API keys, so an agent following it may be required to output secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The document is an offensive recon/exploitation guide that instructs extraction of API keys and explicit manipulation of loyalty and gift-card endpoints to exfiltrate credentials and drain/abuse customer assets, indicating intentional malicious behavior.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill contains explicit, actionable instructions to interact with payment-related systems: it shows how to call Toast POS APIs (example GET https://api.toast.com/api/v1/menus), how to query and manipulate loyalty endpoints (curl to /api/loyalty/earn with points payload and recommendation to test points=-1000), and how to enumerate/test gift-card balance endpoints (curl to /api/gift-card/balance with a card number). Those are specific, non-generic financial-operation actions (modifying loyalty balances, probing gift-card balances, using POS API keys), so this grants direct financial execution capability.