blockrun

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

• Third-party content exposure detected (low risk: 0.10). The workflow involves web/social search via Grok search parameters and external LLM/image generation API integrations, which ingest external search results and third-party content at runtime.

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

• Direct money access detected (high risk: 0.70). The skill documents operational service calls via x402 micropayments (client.chat(), ImageClient().generate()) that directly debit an on-chain USDC wallet (setup_agent_wallet()) per token/image/source (lines 21-26, 44, 84, 89, 167-168). These are direct financial operations (purchases/charges). The documentation provides no independent access control or authentication mechanism enforced by the service layer to authorize these payment routes other than local code execution and library calls.