init-decision-tree

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt tells the agent to detect and include environment variables from .env/.env.example/.env.local into generated outputs (CLAUDE.md) but does not require masking or omitting values, which could force the LLM to read and verbatim-output secrets.

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

• Direct money access detected (high risk: 1.00). The skill explicitly includes SaaS billing options and names a payment gateway: it lists "Payment model? a) Stripe subscriptions → stripe-subscriptions + billing-automation" and "Key integrations needed? a) Stripe (payments)". Those are specific references to a payment gateway integration (Stripe) and billing automation, which qualifies as direct financial execution capability per the rule.