playbook-creator

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

• Third-party content exposure detected (medium risk: 0.30). In the “When to Trigger” execution path, the skill ingests outsider-authored PRD/app-concept free text from the user prompt (“Create a playbook for this PRD”, etc.) and uses that text at runtime to generate the sequential micro-prompts, so the workflow is directly driven by arbitrary user input.

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

• Direct money access detected (high risk: 1.00). The skill explicitly references Stripe and contains specific instructions to implement Stripe checkout and webhook API routes, plus database fields for stripe_customer_id and stripe_subscription_id. Those are concrete payment-gateway integrations (instructions to build code that handles payments/webhooks), which constitute direct financial execution capability.