dailymed-spl-search

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

• Third-party content exposure detected (low risk: 0.10). The runtime workflow calls https://fittings.sh/v1/dailymed/spl-search using user-supplied query parameters (e.g., drugName) but it ingests only that structured input plus first-party DailyMed API responses; it does not ingest arbitrary outsider-authored free text from an external feed without selecting a specific API item.

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

• Potentially malicious external URL detected (high risk: 0.80). The SKILL.md instructs fetching and then running remote code with "curl -sO https://fittings.sh/skill/call.mjs", which is a required runtime external dependency that would download and allow execution of remote script code.

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

• Direct money access detected (high risk: 1.00). The skill explicitly supports on-chain payments for API calls: it returns signed quotes (402), requires the client to sign an authorization, and provides an example where a private key (FITTINGS_PRIVATE_KEY) is set so the client will spend USDC on Base. The docs instruct running the provided client to perform real payments and reference EVM-related libraries (@x402/evm). Because the skill is explicitly designed to initiate and authorize real monetary transfers (signing and causing payment settlement), it grants direct financial execution capability.