google-news

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). This skill includes explicit API-key-like literals and Authorization header examples (including an MCP config with a Bearer token), which encourage or require embedding secret values verbatim in commands/configs and thus risk secret exfiltration.

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

• Direct money access detected (high risk: 1.00). The skill explicitly describes payment mechanisms: prepaid credits via POST /api/credits/topup and a x402 pay-per-call flow that requires USDC on specific blockchains (Base, Polygon, Arbitrum, Monad, Solana) and references @x402/fetch which "signs and retries automatically." It also names a paid operation topup_credits. These are explicit crypto/payment actions (signing and moving on-chain funds / topping up credits), so the skill grants direct financial execution capability.