internet-court

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly states the router may "fetch them on demand from the repo" and instructs loading a sub-skill's SKILL.md "from the repo's raw URL", so runtime fetching from https://github.com/internet-court/internet-court-skill can inject prompt/instruction content that controls agent behavior.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly designed to handle financial transactions and custody: it describes agent-to-agent commerce, payments, escrow, delegated wallets and spending mandates (mentions x402, MPP, ERC-7710/7715, Arkhai/Alkahest escrow), and requires producing transaction evidence (tx hashes, signed receipts). Those are specific payment/crypto/escrow capabilities (not generic browser or API tooling), so it grants direct financial execution authority.