mpp

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt includes explicit examples that embed API keys/secrets inline (e.g., apiKey: 'sk-...', secretKey: '...') and proxy configurations that would encourage copying user-provided keys verbatim into generated code or commands, creating an exfiltration risk despite some secure env-var patterns also being shown.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly a payments SDK/protocol. It defines and exposes concrete payment methods and settlement rails (Tempo stablecoins, Stripe cards, Lightning Bitcoin, Stellar/Card, custom rails), SDK APIs for creating charges and sessions (mppx.charge, mppx.session), on-chain transaction preparation and signing (feeToken, setUserToken, prepareTransactionRequest, signing via viem/private keys, Privy server wallets), integration with Stripe and Lightning SDKs, and examples that construct and submit payments (stripe.charge, tempo({ account }), spark.charge with mnemonic). It also supports payment-gated proxies, MCP tool payments, opening/settling channels, split payments, and a CLI to make paid requests. These are specific, purpose-built financial execution capabilities (payment gateways, crypto wallets, on-chain settlement, and programmatic charge issuance), so it grants Direct Financial Execution Authority.