api-onboarding

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly encourages exposing and pre-populating API keys (including literal test/live examples like "sk_test_abc123" / "sk_live_abc123") and embedding them into examples and copy-pasteable code, which would require the LLM to include secret values verbatim in its output.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes configuring “Interactive Documentation” (e.g., a “Try it” API explorer with editable request parameters and a response viewer), which implies the runtime may ingest user-provided free-text for request inputs and display returned text, creating potential indirect prompt-injection exposure.

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

• Direct money access detected (high risk: 1.00). The skill includes explicit payment/charging examples and test payment data — e.g., client.payments.create(...) calls, test card numbers for successful/declined charges, and "sk_test_" test API keys. These are concrete APIs/examples for creating payments (moving money or simulating charges) rather than generic tooling, so the skill grants direct financial execution capability.