fittings-geo-forward-geocode

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). The skill instructs users to download and execute a JavaScript module (call.mjs) directly from an unverified domain, which is equivalent to running an executable from an untrusted source and is a high-risk distribution vector for malware.

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

• Third-party content exposure detected (medium risk: 0.30). In the geo.forward-geocode workflow at GET https://fittings.sh/v1/geo/forward-geocode, the agent runtime ingests the outsider-provided free-form address string directly as input to the geocoding request.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's instructions show downloading and then executing remote code via "curl -sO https://fittings.sh/skill/call.mjs" followed by "node call.mjs", which fetches and runs external code at runtime (https://fittings.sh/skill/call.mjs).

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

• Direct money access detected (high risk: 1.00). The skill explicitly describes a paid API flow that requires signing an authorization with a private key and using a wallet holding USDC on Base (export FITTINGS_PRIVATE_KEY, signing quotes, facilitator settles). It references EVM-related client libraries (@x402/evm) and instructs how to perform on-chain/off-chain payment authorization and automated spending caps. Those are explicit crypto payment/wallet actions (signing authorizations to spend funds), so this grants direct financial execution capability.