miroshark-matchday

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

• Third-party content exposure detected (low risk: 0.10). The workflow performs web searches and fetches simulation status and reports from miroshark.xyz, ingesting external text from web pages and API responses.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs downloading code and instructions from an external repository (https://www.remotion.dev/docs/ai/skills and https://www.miroshark.xyz/skills/miroshark-video.md) to build and execute rendering logic, introducing high-risk unverifiable external runtime dependencies.

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

• Direct money access detected (high risk: 1.00). The skill includes explicit capabilities and instructions for executing paid transactions (specifically paying $1.00 USDC via x402 on Base using a connected Finance District agent wallet) and includes onchain writes as a declared capability.