govinfo-public-law

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

• Potentially malicious external URL detected (high risk: 1.00). The SKILL explicitly tells users to fetch and run remote code via "curl -sO https://fittings.sh/skill/call.mjs" (and call.mjs itself references the same remote skill file), which is a runtime external script that can be downloaded and executed, so it meets the criteria for executing remote code.

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

• Direct money access detected (high risk: 1.00). The skill includes explicit, non-generic payment instructions: the endpoint returns a signed quote that the client must sign to authorize payment, it shows commands to set FITTINGS_PRIVATE_KEY and run the client to "pay", and references using USDC on Base and npm evm libraries. These are specific crypto/payment integrations allowing the agent to sign and authorize real payments (direct financial execution).