fittings-url-expand

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). These URLs include a URL shortener (bit.ly) that hides the final destination and a direct link to a downloadable/runnable script (https://fittings.sh/skill/call.mjs) with explicit instructions to fetch and execute it—both are high-risk indicators for malware distribution.

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

• Third-party content exposure detected (medium risk: 0.30). In url.expand the required workflow explicitly performs a live fetch/dereference of the caller-supplied url (e.g., shortened links) and follows the redirect chain to the final destination, so outsider-provided free text from that URL is ingested at runtime.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill README explicitly instructs fetching and executing remote code via "curl -sO https://fittings.sh/skill/call.mjs" followed by "node call.mjs", so the URL https://fittings.sh/skill/call.mjs is a runtime external dependency that will execute remote code.

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

• Direct money access detected (high risk: 1.00). The skill includes an explicit crypto payment flow: it returns a signed quote (402), requires the caller to sign an authorization with a private key, and shows using a FITTINGS_PRIVATE_KEY (environment variable), USDC on Base, and @x402/evm tooling. Those are concrete wallet/signing instructions that enable an agent to authorize payments (direct financial execution), not just generic HTTP calls.