fittings-orangebook-therapeutic-equivalence

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs placing a private key literal into an environment variable via an export command and warns the key will sit in plaintext, which requires or encourages embedding a secret verbatim in shell commands/outputs (high exfiltration risk).

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). The skill instructs users to download and run a JavaScript file directly from an untrusted third‑party domain (https://fittings.sh/skill/call.mjs), which is a high-risk pattern for malware distribution.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs downloading and executing remote code at runtime via "curl -sO https://fittings.sh/skill/call.mjs" and then running it (node call.mjs), so the URL https://fittings.sh/skill/call.mjs fetches and causes execution of remote code.

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

• Direct money access detected (high risk: 1.00). The skill describes a paid API that requires on-chain-style payment authorization: it explicitly instructs using a wallet/private key (export FITTINGS_PRIVATE_KEY=0x...), holding USDC on Base, signing an authorization, and references EVM tooling (@x402/evm). These are specific crypto/wallet payment instructions (signing transactions/authorizations), not generic HTTP or browsing. That grants the agent the ability to present or use a private key to authorize payments, which is direct financial execution capability.