agent-agentic-payments

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's toolkit explicitly requires passing sensitive values like private_key_hex and signature_hex as literal parameters in generated calls (e.g., signing mandates), which would force the agent to include secret/private-key material verbatim in its outputs.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly a payment/authorization system and contains specific, finance-focused APIs and actions. It defines functions to create and manage mandates with spend caps, sign mandates with Ed25519 keys, authorize payments (mcp__agentic-payments__authorize_payment), request/verify multi-agent consensus for approvals, revoke mandates, and track payment status. It also states "ACP (Agentic Commerce Protocol): REST API integration with Stripe-compatible checkout" and lists real-world use cases such as e‑commerce purchases, robo-advisors executing trades, and enterprise procurement. These are specific tools for moving/authorizing money and cryptographic transaction signing (crypto/blockchain-related), not generic automation. Therefore it grants direct financial execution authority.