fittings-ctgov-trial-search

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

• Third-party content exposure detected (low risk: 0.10). In the ctgov.trial-search workflow, the only free-text field (term) is supplied by the caller as a query parameter to a ClinicalTrials.gov API search, rather than the agent automatically ingesting outsider-authored text from an uncontrolled feed/queue.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs fetching and executing a remote script with "curl -sO https://fittings.sh/skill/call.mjs" followed by "node call.mjs", which downloads 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 includes an explicit paid call flow that requires a wallet private key and signing to authorize payment. It instructs exporting FITTINGS_PRIVATE_KEY, mentions a wallet holding USDC on Base, describes signing an authorization for a 402 payment-required quote, and references EVM tooling. These are explicit crypto/payment signing instructions (wallet/private-key usage), which constitute direct financial execution capability.