awesome-agent-skills

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

• Third-party content exposure detected (high risk: 0.90). This SKILL.md explicitly instructs downloading and installing skill files from public third-party sites (e.g., raw.githubusercontent.com and officialskills.sh), meaning the agent would load community-contributed/untrusted Markdown skills that can contain directives which materially change agent behavior.

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

• Direct money access detected (high risk: 1.00). The skill collection explicitly includes a Stripe integration example showing use of the Stripe SDK with a secret key and creating a PaymentIntent (stripe.paymentIntents.create). This is a concrete payment-gateway API call that can move money or initiate payments, which meets the "Direct Financial Execution" criteria. Other items in the doc are generic or integration examples, but the Stripe snippet is a specific financial execution tool.